Previous day | Jump to hour: 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 | Next day

Seconds: Show Hide | Joins: Show Hide | View raw
Font: Serif Sans-Serif Monospace | Size: Small Medium Large

Click in the nick column to highlight everything a person has said.
The Logo icon identifies that the person is a core developer (has commit access).

Notice: Only Gecko based browsers prior to FF4 support the multipart/mixed "server push" method used by this log reader to auto-update. Since you do not appear to use such a browser, this page will simply show the current log, and not automatically update.

#rockbox log for 2007-07-13

00:00:02amiconnOtherwise gcc picks the __GNUC__ code below, which adds 1336 bytes to the resulting codec
00:00:23 Quit davina (Remote closed the connection)
00:00:31linuxstbAlthough I'm not sure if it's that big an issue - the supporting files (bitstream.[ch], bswap.h etc) aren't that important. It's wmadeci.c that would be nice to donate back to ffmpeg.
00:00:34stripwaxRbutil Error : Failed to display HTML document in UTF-8 encoding
00:00:37stripwaxnot a great start!
00:00:51stripwax(just hitting Autodetect)
00:01:03 Join My_Sic [0] (
00:01:17amiconnlinuxstb: My changes only touch bswap.h, bitstream.h (single line), ffmpeg_config.h, and now the Makefile
00:02:31stripwaxClicked on install bootloader, got "Unsupported Bootloader install method".
00:02:32linuxstbThen IMO it's not worth worrying about. But saratoga's done most of the work on the decoder itself, I've just been working on the ASF layer and Rockbox integration.
00:02:53stripwaxDetails said "The Bootloader has been installed on your device. Unsupported Bootloader Install method".
00:03:13stripwaxYou're right. It's not really clear what's going on, even if you just click the button
00:03:15 Quit jhulst (Read error: 104 (Connection reset by peer))
00:03:40dan_aAs far as installation instructions go, I would replicate section 2 in its entirety for installation on Windows, Linux and MacOS, and try to give less choices in the install instructions.
00:04:01stripwaxIt also preselects the wrong directory for the connected device. That should be automatic (especially if Autodetection seems to work)
00:04:03linuxstbstripwax; I think we would want to get rbutil to the stage where we just tell users to download it and follow the onscreen instructions.
00:04:10stripwaxyep, definitely
00:05:16linuxstbdan_a: section 2 is the part that talks about the different available builds?
00:05:40stripwaxweird, rbutil is faster at unzipping that windows zip ;-)
00:05:49stripwaxunzipping a build and copying to device that is
00:06:04DanHibikilinuxstb: with the latest build, almost every time i try to boot rockbox i get this .rockbox directory not found message, the few times it booted properly, i tried to play some songs with no luck
00:06:33linuxstbDanHibiki: I assume you're not the only Nano owner using a recent build, so it's very odd...
00:06:47dan_alinuxstb: Section 2 is the entire installation instructions. I know that it's only the bootloader install that differs for the 3 OSs, but the simpler the instructions look, the less scary they will be
00:07:20 Join Wiwie [0] (
00:07:42webguest06Hello again people, i have succesfully installed Rockbox on my Ipod :) now how do i put my music back on it?
00:07:47linuxstbdan_a: OK, so have an "Install from Windows" chapter, an "Install from Mac OS X" chapter, and a "Install from Linux/Unix" chapter.
00:07:48DanHibikilinuxstb: i find it odd too, but it is happening =(
00:08:29stripwaxwebguest06 - copy it
00:08:30DanHibikigot corrupted folder name, recpresets became rdcpresets
00:08:32stripwaxdrag + drop
00:08:47DanHibikii'm gonna try 2007-07-02 now
00:09:00linuxstbwebguest06: Do you still want to use the original Apple firmware on your ipod?
00:09:06webguest06Dont mean to be an idiot, but drag and drop from where? and to where?
00:09:18webguest06I dont know, acan i do that and still play the cool games?
00:10:07stripwaxfrom your PC onto your ipod
00:10:19stripwaxwindows explorer
00:10:21webguest06into itunes? or can i not use itunes anymore?
00:10:25stripwaxor My Computer, or whatever
00:10:45stripwaxif your music was already copied on using itunes, then it's still on your ipod, you don't need to copy it over again
00:10:51stripwaxit's still there
00:11:03webguest06oh ok lol how to i access it?
00:11:11stripwaxDid you read the manual at all?
00:11:22Soul-SlayerSettings -> General Settings -> Database -> Initialise now
00:11:23webguest06when i go database its says Not ready, want to initialize it now?
00:11:24stripwaxActually we get asked all these questions all the time so we wrote the manual to answer them
00:11:27stripwax :-)
00:11:33Soul-SlayerYes, you do.
00:11:41webguest06Oh great thanks
00:11:47Soul-SlayerThen restart when it's done and go into database again, all your old music will be there
00:12:10 Quit linuxstb ("CGI:IRC")
00:12:21webguest06how long will it take approz to initialize? seconds? or hours ? lol
00:12:53stripwaxmaybe one minute or so
00:13:01stripwaxprobably less unless you have a lot of tunes
00:13:13webguest06oh kk its working now
00:13:17stripwaxwebguest06 - what music format are you using? (mp3? aac?)
00:13:26webguest06i have about 1500 tracks alogng with 4 films
00:13:32webguest06mostly mp3 i ithink
00:13:49webguest06im so sorry for the fuss im a complete noob at computers
00:14:19stripwaxwebguest06 - are you sure you're ready for Rockbox? ;-)
00:14:22*stripwax is joking of course
00:15:02webguest06honestly i dont know much about it, just today at school my friend showed me his, :) lots of cool stuff, including game Doom :)
00:15:28amiconnHmm, wma shows the same weird boost behaviour as aac on coldfire :(
00:15:40stripwaxhm, rbutil theme downloading doesn't work, apparently the downloaded zip has "the wrong directory structure".
00:15:47amiconnWhile it's now ~145% realtime, it stays 100% boosted during playback
00:16:00stripwaxit's kinda unfriendly, I'll see what I can do
00:16:00webguest06nice one :D all music works well
00:16:09webguest06just 1 more question
00:16:12stripwaxgo for it
00:16:15 Quit Rincewind ("Cya")
00:16:50webguest06when i click on artist then all tracks they arent in album order
00:16:55webguest06anyway to fix this?
00:17:15 Join SirFunk [0] (
00:17:37webguest06oh okay, i just click on the album isntead of all tracks :D no worries thats great
00:17:50webguest06strip waz and soul slayer thanks to you both :)
00:17:57Soul-SlayerNo problems
00:17:59webguest06wax* lol i swear my x key isnt working
00:18:08stripwaxwebguest06 - if you get a chance, spend some time on the manual, it's pretty good
00:18:09stripwaxno worries
00:18:14amiconnHmm, seems this happens only with voice enabled?? =:\
00:18:19stripwaxyou can call me strip waz if you like
00:18:26Soul-SlayerAnd the wiki if you're feeling adventurous
00:19:06webguest06are there any other cool things to do immediatly or do i have to install other programs to get Doom etc.
00:19:40Soul-SlayerYou need a .wad file to play doom
00:19:54bluebrotherwell, rbutil can install that for you ;-)
00:19:57Soul-SlayerThe shareware version and freedoom are accessable on the site somewhere I beleive
00:20:02stripwaxwebguest06 - again, read the manual, especially the section about plugins and games and applications
00:20:11webguest06ok great thanks again
00:20:19bluebrotherwebguest06: check and
00:20:54 Quit webguest06 ("CGI:IRC")
00:21:39 Join webguest68 [0] (i=5b7d192a@gateway/web/cgi-irc/
00:21:50*stripwax doubletakes
00:21:55webguest68hey guys its me, 06 - closed IRC by accident :P sorry
00:22:12bluebrotherwebguest68: we already noticed ;-)
00:22:29 Quit Arathis ("Bye, bye")
00:22:31amiconnPython tracebacks aren't nice :/
00:22:36*bluebrother likes IP masks
00:22:41 Quit Siltaar_ (Remote closed the connection)
00:23:11*petur wants the webirc client to use a stupid default nickname so more people would select a proper one
00:23:26bluebrothersomething like iamanidiot?
00:23:36peturamiconn: using viewcvs?
00:23:47 Part Infinoid
00:23:52amiconnYes, viewvc is broken atm
00:23:56bluebrotheriamben: oops ;-)
00:23:59*amiconn pings Bagder
00:24:14*Bagder mrfms about bed time
00:24:28Bagderzagor is the poker of all things python
00:24:51peturno hurry, it's been broken all day I think
00:24:54amiconnpython poker?
00:25:16webguest68hey yet another noob question xD im on the manual looking at plugins and i see blackjack at the top but no link to get it
00:25:22*stripwax downloads wxwindows
00:25:23webguest68how do i download it onto the ipod?
00:25:32bluebrotherwebguest68: you don't download it
00:25:34Soul-SlayerIt's already there
00:25:37stripwaxwebguest68 - almost everything in the manual is ALREADY installed!
00:25:37Soul-SlayerThey're all already there
00:25:45bluebrothereverything described in the manual is part of a standard install
00:25:50Soul-SlayerMain menu -> Plugins -> Blackjack
00:25:52stripwaxso just load the game and play it
00:25:53webguest68oh rofl xD where are they hidden :P
00:25:55bluebrotherexcept otherwise noted (like the doom wad)
00:26:01stripwaxMenu.. Plugins...
00:26:06webguest68nice one
00:26:07stripwaxseriously, read the manual ..
00:26:12bluebrotherwell, if you read that chapter carefully it tells you at the top ...
00:26:25stripwaxit's all in there. otherwise we'll be here all night telling you the same as just what's in the manual
00:26:26webguest68my bad
00:26:33*bluebrother wonders how hard it could be to not find that information ...
00:26:55webguest68lol i immediatly went for the list of games down the left side, didnt see top paragraph
00:26:58 Quit DerPapst ("So Long And Thanks For All The Fish!")
00:27:14Soul-SlayerOn a totally unrelated note, could someone like make the gigabeat power calibration slightly more accurate? I keep thinking it's about to die when I have hours left, lol
00:27:19stripwaxbest to start at the top, and work downwards, in that order ;-)
00:27:25bluebrothermight I suggest you spending 2 hours of concentrated reading with the manual and coming back with specific questions afterwards?
00:27:27Soul-SlayerWhich file would handle that anyway?
00:28:29webguest68ok well im off now, cya guys, thanks for help
00:28:55PaulJamSoul-Slayer: do you mean the remaining battery time estimation?
00:29:38Soul-SlayerWell, the % itself, it jumps from like 93% to 70%, and lingers on 0% for ages
00:30:05PaulJamah, ok. then i have no idea.
00:31:03stripwaxwebguest68 - happy reading
00:31:03 Quit webguest68 ("CGI:IRC (EOF)")
00:31:54peturSoul-Slayer: firmware/powermgmt.c would be a good place to start looking I think
00:32:34Soul-SlayerVery true
00:34:45 Join midkay [0] (n=midkay@rockbox/developer/midkay)
00:36:33amiconnSoul-Slayer: If this happens, the scale needs calibration. You can help by running a battery runtime test as described here:
00:36:52 Join Robin0800 [0] (
00:36:57 Join BigMac [0] (
00:36:58Soul-SlayerBeen and done
00:37:02Soul-SlayerAdded it to the wiki
00:37:08amiconnThen the table for percentage calculation in powermgmt.c can be calibrated with the result
00:37:50Soul-Slayer Last entry was mine
00:40:02peturhmmm that voltage barely changed the last two hours
00:40:07 Quit _Veseliq_ (Remote closed the connection)
00:40:14 Join Soul-Slaye1 [0] (n=Administ@
00:41:12 Join bdgraue_ [0] (
00:41:48Soul-Slaye1Not sure where my internet cut out then, but it stays on 0% for 2 hours or so and skips from 90 to 70 :$
00:42:52peturthe voltage barely changed the last two hours
00:43:49Soul-Slaye1Kinda begs the question -why-
00:43:53 Quit ender` (" A bus station is where the bus stops. A train station is where the train stops. On my desk, I have a workstation.")
00:44:39 Quit ompaul ("Leaving")
00:44:49*stripwax needs to sleep, wxwindows takes way to long to build
00:44:52stripwaxbye folks
00:45:14Soul-Slaye1Sleep well
00:45:20 Join bdgraue__ [0] (
00:45:53dan_aHow would people feel about moving the installation instructions to something like - the idea is to keep the existing clarity, but give less details of the options available. This makes the instructions shorter, and hopefully less scary looking
00:48:49 Join Dwyloc [0] (i=53688643@gateway/web/cgi-irc/
00:49:23FebsI think that the installation instructions will be completely re-written once Rockbox Utility becomes the official installer, and they should be much more user friendly.
00:49:48bluebrotheryeah ... rbutil will cause some changes.
00:50:05dan_aOf course
00:50:28bluebrotherbut I was thinking about some "expers" section −− maybe we should move them to some "experts only −− manual installation" section when time comes?
00:50:45Dwylocamiconn: On the topic of battery meter calibration any reason why has not been committed yet?
00:51:14DwylocAs it helps quite alot on ipods.
00:52:03dan_abluebrother: Why would people want to do a manual installation? (I've not used rbutil yet, so there could be really obvious reasons)
00:52:35bluebrotherI'd just call it "more control" −− rbutil hides some things
00:54:35dan_ahmmm... I ought to have a look at it, I guess. What sort of things does it hide? Installing arbitrary bootloaders?
00:55:05 Quit obo ("bye")
00:55:23 Quit Robin0800 (" HydraIRC -> <- Like it? Visit #hydrairc on EFNet")
00:55:30 Join Robin0800 [0] (
00:55:50 Quit secleinteer (Remote closed the connection)
00:56:03peturDwyloc: so the values that patch introduces are ok?
00:56:41 Quit bdgraue (No route to host)
00:57:45DwylocThey seem to work quite well on my ipod video 80GB and my nano 1st gen.
00:58:15 Quit Soul-Slayer (Read error: 110 (Connection timed out))
00:59:13DwylocI dont have a 30GB or 60GB ipod to test on though, so I cant coment on how well it works on them.
01:00:01 Join jhulst [0] (
01:00:45 Quit bdgraue_ (No route to host)
01:01:23Robin0800whats the patch number?
01:02:13 Join Freakingme [0] (
01:02:28Freakingmehow stable is rockbox for the cowon Iaudio x5?
01:02:32bluebrotherhmm, that patch contains tabs :o
01:03:10Soul-Slaye1If it has an official build, it's stable
01:03:33***Saving seen data "./dancer.seen"
01:03:50 Quit bdgraue__ (No route to host)
01:03:50Robin0800petur,Ive used this patch on a 60GB Ipod it seems to work very well and gives no problems
01:03:52peturFreakingme: x5 is well supported...
01:04:03 Quit Soul-Slaye1 (Read error: 104 (Connection reset by peer))
01:04:26bluebrotherit's pretty stable.
01:04:45Freakingmetnx. Before I try, I need to know one more thing, is it kinda easy to install back the original firmware?
01:05:30peturI think so yes. X5 is however the only one that doesn't offer dual-boot
01:06:35amiconnM5 is the same, afaik the gigabeat also doesn't dualboot.
01:06:39 Join Soul-Slayer [0] (n=jonno@
01:06:51 Join Tr1ckY^Gone [0] (
01:06:55 Nick Tr1ckY^Gone is now known as Tr1ckY (
01:07:24Tr1ckYwhat to do if rockbox crashes on iPod 5.5G ?
01:07:35PaulJamTr1ckY: reset it
01:07:42Tr1ckYPaulJam, how ?
01:08:00PaulJami think by holdung menu + select
01:08:07iambenremove the batte... oh nm
01:09:05Tr1ckYthat worked, PaulJam... the same thing however, didn't seem to respond before
01:10:42Tr1ckYthanks for the help :)
01:10:47PaulJamTr1ckY: i think i read somewhere that it is sometimes required to flip the hold switch on and off before holding the buttons
01:10:57 Join Nico_P [0] (
01:11:13Nico_PLlorean: hi
01:11:48bluebrotherTr1ckY: also, you need to hold the buttons for a while. The time seems to vary, and can be something up to 40 seconds or like that.
01:12:15Tr1ckYthat's a long time
01:12:46bluebrotheryeah. But that's a feature apple added so nothing Rockbox can do about
01:12:57bluebrotherbut it's really nice to have that feature ;-)
01:13:09 Quit Wiwie ("Miranda IM! Smaller, Faster, Easier.")
01:14:56 Quit petur ("Zzzzzz")
01:16:07 Join sarixe [0] (
01:17:11 Join RogerBacon [0] (
01:19:40Tr1ckYhehe bluebrother
01:19:55Tr1ckYi just made it through the first leve in FreeDoom :P
01:21:12 Quit BigMac ("Ex-Chat")
01:24:52 Quit Robin0800 (Read error: 104 (Connection reset by peer))
01:26:10 Quit Dwyloc ("CGI:IRC (EOF)")
01:26:45 Quit Entasis (Read error: 104 (Connection reset by peer))
01:27:15 Join bdgraue [0] (
01:28:17 Join BigMac [0] (
01:29:56 Quit PaulJam (".")
01:31:54 Quit RogerBacon ()
01:32:08 Quit miepchen^schlaf ("Verlassend")
01:34:51Nico_PLlorean: just wanted to tell you I'm on vacation for a fex days... back friday evening or saturday morning
01:35:25Nico_PI didn't really get a chance to come here before today
01:35:32 Quit atsea- (Remote closed the connection)
01:36:35 Quit Nico_P (Remote closed the connection)
01:36:44 Quit LqR ("sl33p")
01:38:23 Quit merbanan ("Leaving")
01:38:55 Join merbanan [0] (n=banan@
01:40:43 Quit jhMikeS (Read error: 104 (Connection reset by peer))
01:41:32 Join jhMikeS [0] (n=jethead7@rockbox/developer/jhMikeS)
01:42:31 Quit kfazz ("Leaving")
01:43:15 Quit entheh ("^~")
01:43:25 Part maffe
01:45:45 Quit Mouser_X (Read error: 104 (Connection reset by peer))
01:45:54 Join Mouser_X [0] (
01:53:47Soul-SlayerWhats the fastest way to waste battery power? 200% pitch, maxed equaliser, high bitrate file and playing a plugin of some sort?
01:58:09 Quit My_Sic (No route to host)
01:58:11zemax volume and constant-on backlight too
01:58:30Soul-SlayerAhh yes, thanks
01:58:59pixelmakeep the disk spinning I'd say (browsing with dircache off maybe)
01:59:03Soul-SlayerWant to see if my attempts at calibrating are any better but I'm also impatient
01:59:22zepixelma: ah yeah
01:59:39Soul-SlayerGood idea, good idea :p
02:00:00zeplaying like flac or wav's help with that too
02:00:03 Join Luis [0] (i=Journey@
02:00:48 Join saratoga [0] (i=9803c6dd@gateway/web/cgi-irc/
02:01:20 Join atsea- [0] (i=ariel@gateway/tor/x-69ff066915342f29)
02:01:20 Join stripwax_ [0] (
02:01:22Soul-SlayerIf only it was this easy to preserve battery =p
02:01:28Luismy 5.5g IPod just started shutting down and hanging randomly while playing music or videos, both in rockbox and in the Apple OS
02:02:04Soul-SlayerFull charge?
02:02:19LuisI'm beinf forced to reset it several times before it starts working again
02:02:48Luisand then it hangs again
02:06:34 Quit Mouser_X (Nick collision from services.)
02:06:45 Join Mouser_X [0] (
02:07:39 Quit Xerion (" ")
02:08:13 Quit Soul-Slayer ("Leaving.")
02:08:47HellDragonrockbox's API is evil
02:09:25 Quit idnar (Nick collision from services.)
02:09:28 Join idnar_ [0] (i=mithrand@unaffiliated/idnar)
02:10:50TiMiD[FD]why ?
02:11:27 Part dan_a
02:11:32HellDragonbecause there's no way you can code anything with it
02:11:38HellDragonexcept if you study it for 10 years
02:12:59midkayHellDragon: rockbox has only been around for half that, and many people can code for it. are you just a very slow learner?
02:13:25HellDragonnot really
02:13:34HellDragonand it was scorche's idea of trolling about it here
02:14:03scorcheno..i was just sick of you whining to me about it in PM, so i said to go talk to others if you think it is
02:14:25midkaytrolling's not a nice thing to do in any case.
02:14:34midkayi suggest that, if it's going to take ten years, you get started soon. :)
02:14:51saratogai don't think the API is bad, but updating the docs would be nice
02:14:58saratogathe docs folder in particular is evil
02:15:03HellDragonsaratoga: the API is bad compared to something like ipodlinux
02:15:14saratogalots of the stuff it suggests doesn't even exist
02:15:36midkayHellDragon: you could always code for ipodlinux then.
02:15:39saratogaHellDragon: sure, but rockbox is tremendiously more lightweight then ulinux
02:15:44HellDragonmaking a _working_ API is common sense
02:15:46HellDragonsaratoga: not raelly
02:15:52midkayi see no problems with the rockbox api at all. our api works just fine.
02:16:02HellDragonmidgey: did you even try coding with it?
02:16:03HellDragonI did
02:16:04midkayit sounds like the only problem is you're not willing to learn it, or you're incapable.
02:16:08midkayHellDragon: i'm a developer.
02:16:13HellDragonthen fix it
02:16:23midkayyou can't fix something that works perfectly fine.
02:16:32scorcheyou have to tell us what is wrong with it first..
02:16:43HellDragonscorche: mostly everything
02:16:48saratogaclearly the API works, given the large number of plugins available and codecs, its jsut not obvious how to do a lot of stuff without spending some time with it
02:17:34 Join Transience [0] (
02:17:58midkaysaratoga: i think most of it makes perfect sense, and anyone who calls themselves a developer should understand how to find the functions in order to see how they work, if necessary.
02:18:15midkayif you have a specific question, feel free to ask. saying "the API sucks, fix it" helps absolutely nobody with anything.
02:18:47scorcheHellDragon: if it doesnt work, then how did we get all fo the many plugins we currently have?
02:19:22HellDragonscorche: yeah, 20k lines of code for a pong is BAD
02:19:57midkayHellDragon: you're just lying and trolling. that's enough.
02:20:09HellDragonmidgey: no because scorche is the one who does the bans
02:20:22HellDragonmidgey: you lied to me, you are the enemy
02:20:23scorcheHellDragon: if you get out of hand, i will ban you
02:20:39HellDragonscorche: you're the one who told me to trll
02:20:43scorchei simply told you that if you had something against the api, bring it up with people here
02:20:44HellDragoni'm just doing what you asked me
02:20:48midkayscorche: i'd say he already has, please do.
02:20:48HellDragoni did
02:20:51saratogaHellDragon: pong is 400 lines not 20000
02:21:13pixelmaXavierGr: still around?
02:21:30saratogaand 1/4 of those are just comments and #defines it uses for clarity
02:22:27midgeyHellDragon: just for the record, you are talking to midkay, not myself. our nicks tend to get mixed up ;)
02:22:40 Quit stripwax (Read error: 110 (Connection timed out))
02:22:42 Join Bspec [0] (i=cider@unaffiliated/bspec)
02:22:58Bspecok, HellDragon, whine away
02:22:58scorcheBspec: you amuse me
02:23:04gregjfolks, since yesterday - I started to get 'undefined instruction at' crash again
02:23:04saratogaactually, just 27 lines of pong.c even use the API
02:23:07HellDragonBspec: no
02:23:10HellDragonBspec: i already did
02:23:11gregjwith 403
02:23:11gregjgcc 403
02:23:18gregjon ipod G4/photo
02:23:46scorchegregj: that shouldnt sure you are using the actual 4.0.3 arm-elf-gcc?
02:24:03 Join Soul-Slayer [0] (n=Administ@
02:24:21scorchehow did you get the compiler?
02:24:27gregjsource code
02:24:33gregjcompiled, installed
02:24:47Soul-SlayerOkay well I worked that out far too linear... Attempt #2!
02:25:01scorchecan you run and see if you still get the error after?
02:25:07 Part Bspec ("i'm not amusing, scorche.")
02:25:12saratogadoes it happen with official builds?
02:25:33 Join Whopper [0] (i=whopper@about/cooking/sandwich/whopper)
02:26:07gregjsure thing shrefiff
02:26:21gregjscorche: i will, hold on
02:26:43scorchegregj: it should take a while, so i dont expect you to come back in the next 20 min =)
02:26:44gregjI had no idea you guys have a script that does all that crap that ihad to do manualy
02:26:54 Join devon [0] (
02:26:55scorcheyeah...our script makes things much easier =)
02:27:01saratogaa script and ready made VMware system too
02:27:19scorcheand colinux
02:29:11devonanybody who has the rhapsody bl with the key signing check patched out?
02:29:28scorcheBagder should be sleeping now
02:29:59saratogadevon: thats possible now ?
02:30:18scorchesaratoga: not afaik..
02:30:29scorchewell, it depends what he is saying..
02:30:58devonbagder wanted someone to change a byte in teh bl
02:31:53devonand mentioned the bl I asked for
02:32:00 Join IOM [0] (
02:32:15 Quit Thundercloud (Remote closed the connection)
02:32:37 Quit Seed (Nick collision from services.)
02:32:45 Join Seed [0] (
02:33:16saratogaoh yes I remember that
02:33:29saratogaas i recall the problem was that no one know how to get the patched bootloader in
02:33:33 Quit Freakingme (Remote closed the connection)
02:33:56devonI may have done so...
02:34:28devonremove battery... go to manufacturing mode... then use e200tool to write...
02:34:35devonI only did the one byte tho
02:35:09saratogadevon: sounds interesting, just becare you don't brick that player
02:35:29devonnope... it still works
02:35:45devonI'm unduly adventurous, I suppose
02:36:01devondig rockbox
02:36:11devonput it on my daughter's nano
02:36:27devonnow I must have it on my 'lil monster
02:37:12devonbut I wanted to try that bl bagder talked about
02:37:42scorchedevon: he should be back in around 6-7 hours
02:38:01devonk... thanks
02:40:57 Join ivan [0] (
02:42:05 Part ivan
02:43:23 Part IOM
02:49:12saratogadoes anyone know whos Rockbox's GSOC admin?
02:49:25scorchebagder is
02:49:55scorcheactually, i think there were a couple of them he promoted, but i am not sure on the technicalities of that
02:50:19scorcheLlorean was one promoted, i believe (but as i said, i dont knwo exactly what that entails)
02:50:40devonapologies... I spoke to soon...
02:50:57saratogai'll assume bagder
02:51:12saratogaBagder: if preglow doesn't fill out my midterm evaluation tomorrow, could you please do it?
02:51:42saratogathe rules say the admin can fill out evaluations if a mentor is out of rearch
02:51:54saratogaI'm not really sure when he'll be around a computer again
02:51:58*scorche watches from above as Bagder's head is gently lying on his pillow
02:53:36 Quit lostnihilist ("Leaving")
02:53:57Transienceis anyone here working on the ZVM port?
02:55:30scorcheTransience: as i told you before, they dont really hang out here...
02:56:53 Quit bdgraue (Read error: 104 (Connection reset by peer))
02:57:34 Join maffe [0] (
02:57:52 Part Transience
03:00:21 Quit Echelon ()
03:02:07devonI see were I made my mistake...
03:02:41devonI read the bl before leaving manufacturing mode...
03:03:01devonback to the drawing board
03:03:35***Saving seen data "./dancer.seen"
03:04:09 Join jac0b [0] (
03:04:57jac0bI want to add a them to the wiki
03:05:18jac0bI registered but now I need write access
03:05:52saratogajac0b: let me try
03:06:00saratogawhats your username?
03:06:19saratogaoh nevermind see
03:09:10gregjsame thing with version that your script compiled
03:09:12saratogaok think i got it
03:09:24gregjon photo ipod: it is always undefined instruction at 00058D60
03:09:40*gregj gets build from site, just in case
03:09:51jac0bsaratoga: do I click on "edit" or "attach"
03:09:56saratogagregj: try the latest SVN build and bootloader
03:10:01saratogaprobably edit
03:10:08saratogai think that lets you add text and files
03:10:21saratogashould probably say what you've attached and all that
03:10:34gregjboodloader is the latest one, unless it got changed within last 2 weeks
03:10:47gregjit is the latest svn
03:10:54gregj revision 13865
03:14:58XavierGrpixelma: I am now
03:16:43 Quit devon ("using sirc version 2.211+KSIRC/1.3.12")
03:18:28 Join rwa [0] (i=cda1d014@gateway/web/cgi-irc/
03:20:10rwai have had rockbox on my ipod video 30gb for about a year now. i copied over the .rockbox folder from the build that i downloaded today and now when i boot the ipod it gives me this error: Data abort at 00030110 - does anyone know what this means?
03:20:51Soul-SlayerDoes it happen all the time or just occasionally?
03:21:16 Join tac0 [0] (i=4c6e5c2c@gateway/web/cgi-irc/
03:21:25rwai just copied the .rockbox folder to it about 10 minutes ago. the ipod will not shut off either and the displays brightness is at 100% and doesnt shut off
03:21:29rwait just displays the error
03:21:42tac0are there screenshots of what the ipod 4g will look like with rockbox on it anywhere?
03:22:02saratogatac0: check the WPS gallery in the wiki
03:22:03Soul-SlayerReset it the usual way, see if it occurs again
03:22:29rwayes i reset it and it gets to the main menu and then immediately shows that error
03:22:46Soul-SlayerThis is the latest build?
03:23:09pixelmawhat error is that?
03:23:19Soul-SlayerA data abort
03:23:27saratogarwa: someone reported a similar problem on the 4G a couple minutes ago too
03:23:31rwaData abort at 00030110
03:23:40pixelmasorry... late here
03:24:07saratogarwa: can you try a build from yesterday and see if it works?
03:24:08rwano worries. do you guys recommend that i try an older build?
03:24:21rwaok sure, one moment
03:26:39Soul-SlayerGrr, why do these battery readings fluctuate so much
03:26:39Tr1ckYis it somehow possible to install iPodLinux on 5.5G iPods ?
03:27:05Soul-SlayerThis channel is for rockbox only
03:27:17Soul-SlayerTry their channel ^^
03:27:53jac0bsaratoga: thanks for the help
03:27:59jac0bheres my theme
03:29:48rwasorry it took so long guys, was having trouble getting my computer to recognize it. i tried yesterdays build but now i get Data abort at 000303F8
03:29:54 Part toffe82
03:31:47Soul-SlayerYou say it happens after getting to the menu?
03:31:51Soul-SlayerOr before?
03:31:58rwaright after
03:32:05rwai can navigate for a split second
03:32:07saratogatry a build from before the menu changes
03:32:17Soul-SlayerHmm... Then try about 2-3 days ago if possible
03:32:41Soul-Slayerjdgordan has played around with menu settings... Seems a possibility
03:32:51 Quit jac0b ()
03:32:52 Part pixelma
03:32:52rwaok ill try the 10th
03:33:26Soul-SlayerShould do it, first menu change seems to be the 11th
03:34:02 Join JdGordon [0] (n=jonno@rockbox/developer/JdGordon)
03:34:15Soul-SlayerSpeak of the devil :p
03:34:17 Quit XavierGr ("One firmware to rule them all!")
03:34:54 Part Llorean
03:35:05JdGordonwhat have i done?
03:35:23Soul-SlayerLol, nothing confirmed until rwa gives feedback on the build he's about to try
03:35:26rwaok the 10th build, it boots to the menu fine, but as soon as i scroll the wheel at all it locks up
03:35:38rwabut no error
03:36:16Soul-SlayerWhich build were you using before?
03:36:48imbecilehow do i play doom on a 5g ipod?
03:36:49rwaoh man, im not really sure, it was from about a year ago - i never updated it because i never had problems. i just thought id give one of the new builds a try
03:36:53 Join Nick_Brackley [0] (i=3aa67da1@gateway/web/cgi-irc/
03:37:00Soul-SlayerTypical that
03:37:03rwaso it would be pretty hard to track down which build it started happening at i suppose
03:37:05rwayeah, sorry
03:37:06Soul-SlayerIn which case
03:37:19Soul-SlayerI trust you reinstalled the bootloader? =$
03:37:21 Quit bluet (Remote closed the connection)
03:37:49rwai have not!
03:37:56saratogadan_a added dual cpu support to the bootloader a couple months back
03:38:08Soul-SlayerThen I suggest you get the latest ipodpatcher, uninstall your current and reinstall the new one
03:38:09saratogagoing to want to have that before trying to run a 2 CPU build
03:38:21Soul-SlayerAlot has happened in a year :P
03:38:41rwahah i guess so, but thats good news.
03:38:53rwawill i find instructions on how to install the bootloader on the site?
03:39:01Soul-SlayerI'll get you a link, hold on.
03:39:07Soul-SlayerAre you using Windows?
03:39:10rwagreat thanks a lot. yeah, xp
03:39:35 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
03:40:15Soul-SlayerRead the instructions on 'Bootloader installation from Windows'
03:40:44 Join DC1 [0] (
03:40:49imbecileanyone have doom working on a 5g ipod video?
03:41:03rwais there a new firmware also?
03:41:07Soul-SlayerI don't, but I may be able to help regardless, what is the problem?
03:41:15imbecilei cant figure out the controls.. nothing happens
03:41:22Soul-SlayerNew firmware just means new build
03:41:30rwacool. thanks a lot for your help
03:41:33linuxstbimbecile: Turn the hold switch on, that brings up the doom menu.
03:41:37Soul-SlayerNo problem
03:41:40linuxstb(then turn it off to use the menu...)
03:41:51Soul-SlayerMake sure it all works before you thank me though :p
03:42:11rwahah, ill try the bootloader right now and let you know what happens
03:43:50tac0assuming i mess up installing this on my ipod 4g grey or just want to revert, how would i do that?
03:44:05tac0the itunes ipod firmware updater?
03:44:32Soul-SlayerRun ipodpatcher again and hit 'u' instead of 'i'
03:44:46Soul-SlayerUninstalls the bootloader, meaning it boots into the Apple OF as usual
03:44:52tac0cool :D
03:46:06rwaSoul-Slayer: it works! thanks a lot. everything was very easy to set up and you guys were very helpful
03:46:21 Quit spiorf (Remote closed the connection)
03:46:22Soul-SlayerNot a problem, go ahead and use the newest build now :p
03:46:44rwaalready am
03:47:04rwai was panicing about this, i have a 4 hour road trip tomorrow and i thought my ipod was screwed
03:47:23Soul-SlayerAslong as you can access it in USB mode there's some hope :P
03:47:27rwai also shorted my cigarette lighter plugin in my car today
03:47:34linuxstbgregj: I've just built and installed current SVN for my ipod Photo and it works fine. Are you using the offiicially released bootloader (the one in ipodpatcher) or a bootloader you built yourself?
03:47:40rwayeah, bad day for the ipod
03:47:42Soul-SlayerI thought I'd busted my gigabeat earlier, lol
03:47:51linuxstbIt is Friday 13th...
03:47:54Soul-SlayerBootloader went into panic every time
03:47:57linuxstb(here, at least...)
03:48:03Soul-SlayerAck! So it is
03:48:07rwaoh didnt even notice that
03:48:29Soul-SlayerHad to open up my gigabeat and hotplug it... Not what I really wanted to do but got the job done
03:48:29rwaalright, im out of here. thanks again for all the help
03:48:34Soul-SlayerSee you, enjoy :0
03:48:42 Quit rwa ("CGI:IRC")
03:49:02saratogalinuxstb: did you put a check to reject V1 files in the asf parser?
03:49:25Tr1ckYSoul-Slayer, what is the name of their channel ? ^^
03:49:26linuxstb(not knowingly...)
03:49:56Tr1ckYnvm, found it
03:49:58saratogamy v1 file gives "incompatible version" in the uisim
03:50:03Soul-Slayer#ipodlinux lol
03:50:19Tr1ckYSoul-Slayer, as i wrote before, i found it
03:50:41Soul-SlayerAnyway... Back to averaging out battery lives... Yay...
03:50:50linuxstbsaratoga: That normally means your wma.codec file is out of sync with your main Rockbox binary.
03:50:57saratogaoh ok
03:50:59JdGordonhas Nico_P been on recently?
03:51:19Soul-SlayerNot within the last hour and a half I've been here
03:55:08 Join mike_ [0] (
03:55:42 Quit mike_ (Client Quit)
03:59:41saratogalinuxstb: my v1 file errors out in the parser
04:00:02saratoga"Could not read packet (3200 bytes), aborting"
04:00:07TiMiD[FD]would someone mind to increase the stack size of 0x100 bytes for the iriver ? I have SVN access no more
04:00:10saratogayou interested in looking at it?
04:00:33TiMiD[FD]that fixes the stack overflow bug in the mpegplayer
04:02:59linuxstbTiMiD[FD]: That isn't necessarily the best solution...
04:03:15linuxstbsaratoga: I'll have a quick look, but it's almost 3am here now...
04:04:03saratogayeah i realize it must be late
04:04:16saratogathe call to request_buffer doesn't give enough memory oddly enough
04:04:25TiMiD[FD]well ... it's just that on the h1x0 the player uses the grayscale lib, which makes it eats more memory than on the color targets
04:04:55linuxstbsaratoga: Yes, I'm just looking at it now. I can't understand why it wouldn't give enough data - assuming it's there...
04:05:40saratogait returns 3186 bytes
04:05:43saratogaat least in the UI sim
04:06:27linuxstbsaratoga: BTW, you know the build table is yellow?
04:06:43saratogayeah i fixed it a couple minutes ago
04:09:36JdGordonis everyone still in favour of ?
04:10:25Soul-SlayerI'm sure I would be if I knew what I was looking at :$
04:10:46TiMiD[FD]ugh that looks like an ipod
04:10:52Soul-SlayerAhh I see the difference
04:10:59Soul-SlayerChanging settings without going to a new menu?
04:11:17imbecileanyone have doom working on a 5g ipod video? i cant figure out the controls
04:11:40Soul-SlayerWell, it would be more practical in my opinion
04:11:43linuxstbJdGordon: I thought there were lots of objections to it...
04:11:56tac0btw- thanks for answering my questions
04:12:01tac0i got it installed and i like it :D
04:12:01Soul-SlayerCoding or aesthetic objections?
04:12:04JdGordonim going through the thread any everyone seems to like it
04:12:08TiMiD[FD]JdGordon: for yes/no settings it's ok, but for those with more options, it's not very convenient
04:12:10Soul-SlayerGreat tac0 :) Enjoy
04:12:18linuxstbJdGordon: Check the IRC logs, not the forums...
04:12:35JdGordonyeah, but forums are easier to find
04:12:50linuxstbYes, but most developers don't discuss things there...
04:13:03TiMiD[FD]and it divides by 2 the numbers of settigns you can see
04:13:59TiMiD[FD]else it looks cute
04:14:36saratogaah i see why mpegplayer has problems with the stack
04:14:48saratogahow much of IRAM does the stack occupy anyway?
04:14:55linuxstbAbout 9KB
04:15:05TiMiD[FD]0x2000 bytes exactly
04:15:09TiMiD[FD]for the main stack
04:15:22linuxstbAh yes, 9KB is the codec stack.
04:15:31linuxstb8KB for the main stack.
04:15:56saratogawhere does the other 40k of reserved IRAM go?
04:16:11linuxstbYou mean 31KB ?
04:16:23linuxstbThe two stacks I mentioned are both in IRAM.
04:16:25tac0 - wheres the download link? :|
04:16:29saratogaoh ok
04:16:44linuxstbtac0: All the plugins are included with a standard Rockbox download.
04:17:34saratogaand on the 128k targets, theres another 32k of IRAM unused right?
04:17:55linuxstbsaratoga: I added ci->curpos to the DEBUGF and it is 2019618 when it aborts - so for some reason it's reaching the end of the file very quickly.
04:18:36saratogalinuxstb: for what its worth, that file was encoded using ffmpeg so its likely the file structure is atypical
04:19:14linuxstbAh, it's the same file I've been testing with...
04:19:25saratogaanyway, that file has been on my back burner for almost a week
04:19:31saratogaits not really a priority for me
04:20:30 Nick Tr1ckY is now known as Tr1ckY^Gone (
04:21:19 Join davidb [0] (i=3ff6afaa@gateway/web/cgi-irc/
04:21:41imbecileanyone have doom working on a 5g ipod video? i cant figure out the controls
04:21:43imbecileanyone have doom working on a 5g ipod video? i cant figure out the controls
04:21:56TiMiD[FD]imbécile ^^
04:22:14imbecileyeah I didnt try to
04:23:08Soul-SlayerAnd surely a small amount of experimentation could help you work out what the controls are, there are only like 10 buttons
04:23:25 Join aaronw [0] (
04:23:28linuxstbAnd I suggested a solution when you first asked the question 45 minutes ago
04:23:34*JdGordon found the relevant irc logs, and works on other stuff today
04:23:53imbecileSoul-Slayer, I only have 2 buttons
04:24:01 Quit aaronw (Client Quit)
04:24:23Soul-SlayerHave you installed rockbox on a mouse or something?
04:25:04 Quit jhulst (Read error: 104 (Connection reset by peer))
04:25:28imbecilelinuxstb, I look to get highlighted for an answer after a few minutes so i dont have to stare at people talking about stuff that doesnt pertain to me then never answer
04:25:55tac0eek... my ipod just said i had a low battery. plugged it in and its not charging or turning on (it had 70% battery 2 minutes ago)
04:26:21linuxstbimbecile: Huh? I quoted your name in my answer, which was 18 seconds after you asked the question...
04:26:44imbecileSoul-Slayer, i realize that ipod wheels have different controls where you push them but technically is 2 buttons
04:27:06imbecilelinuxstb, what time was it?
04:27:28Soul-SlayerMenu, Select, Fast Forward, Play, Rewind, Wheel Left, Wheel Right, Hold
04:27:37linuxstbimbecile: Read my answer from 3 minutes ago...
04:27:39*linuxstb sleeps
04:27:45imbecileill go through logs cuz my xchat hasnt flahed in ages
04:28:06imbecileno timestamping?
04:28:08Soul-SlayerAnd I gave you a link to the wiki page saying all the buttons.
04:28:25Soul-SlayerScroll down to keys, look at the heading 'iPod'
04:28:32linuxstbimbecile: Search for your nick in the log -
04:28:33imbecileSoul-Slayer, i know i got it.. thanks
04:28:45Soul-SlayerSleep well linuxstb
04:29:05imbecilelinuxstb, I save logs myself thats why i was asking
04:30:02imbecilelinuxstb, ahh i see where you said it.. you were right, sorry
04:31:18 Quit tac0 ("CGI:IRC (EOF)")
04:31:19imbecilexchat usually blinks on me when i am highlighted
04:36:25 Quit saratoga ("CGI:IRC")
04:37:26 Quit SirFunk (Connection timed out)
04:38:24 Join SirFunk [0] (
04:42:09 Join aliask [0] (
04:42:39 Quit Tr1ckY^Gone (Read error: 110 (Connection timed out))
04:43:08gregjwell, bad news everyone - the one I downlaoded from your website, is also borked - crashes as well :(
04:43:47 Join Tr1ckY^Gone [0] (
04:55:58 Join Justi1 [0] (
04:56:09Justi1does rockbox play .ogg files?
04:58:07sarixeJusti1 : of course
04:58:23 Join Llorean [0] (
04:58:37 Part Justi1
05:03:37***Saving seen data "./dancer.seen"
05:04:28Nick_Brackleyhas anyone been experiencing problems with mpegplayer? i try to play the elephants dream sample movie on my iriver h300 and get "*panic* stkov main"
05:05:07imbecilejust curious of how things are setup in rockbox.. is there a "by artist" or "by album" option? the reason i ask is i have about 20gb of files in one folder and i dont want to have to scroll though all kinds of stuff to find what i'm looking for
05:06:01Nick_Brackleyyes there is in the database (tagcache)
05:06:03LloreanNick_Brackley: I believe at the moment mpegplayer crashes when creating its config file. If you manually create an empty file of the right name in its folder, it should be fine. I don't know the name though
05:06:13LloreanAlso, it's not called "tagcache" any more.
05:07:27Nick_BrackleyLlorean: Ok will try it, and i wasnt sure so i thought i'd put both down... :)
05:13:48 Part Whopper
05:14:14 Quit DC1 ("$4E75")
05:23:09Nick_Brackleyhmmm cant seem to find any references of where the config file needs to go :(
05:24:00LloreanJust in the same folder as mpegplayer, I believe
05:30:22imbecileis rockbox its own type of os or is it based on linux?
05:30:42imbecilethat question wasnt really asked right
05:31:05imbecilei was just wondering if it was linux based
05:31:37Nick_Brackleyphrased that wrong, i also dont know what it needs to be named
05:31:58Doomedin the root of the .rockbox folder
05:32:22 Quit davidb ("CGI:IRC (EOF)")
05:34:54Soul-Slayerimbecile: Rockbox is all original code, it's not linux based.
05:35:31Soul-SlayerAs the main page says: "Rockbox is a complete rewrite and uses no fragments of any original firmwares."
05:35:48LloreanSoul-Slayer: Well, that wouldn't rule out Linu
05:35:58LloreanBut yeah, 'tis not Linux based.
05:36:04Soul-SlayerEhh well, it isn't linux based, so there we go :p
05:37:39imbecileok thanks
05:38:17imbecileanother question, what is the diff of battery life compared to appleos?
05:38:28Doomeda lot, check the wiki
05:38:46Soul-SlayerLess, all portalplayer chips have a battery life problem apparently
05:40:08Soul-SlayerAs you can see from the few entries that have filled in the OF time, it's about half
05:42:02imbecileyeah, oh well i can use rockbox when i can charge but on trips etc.. ill just use appleos
05:42:24imbecilei love the look and feel of rockbox tho
05:43:59Soul-SlayerIf you plan on switching between the two, make sure you transfer music using iTunes or the OF won't be able to pick it up
05:51:31 Join bonbonthejon [0] (
05:58:27 Quit sarixe ("Peace")
05:59:01 Quit bonbonthejon (Remote closed the connection)
06:01:09 Part Soul-Slayer
06:01:23imbecileSoul-Slayer, yeah I did that.. took me 5 hours to transfer just 12gb
06:04:28 Join ubuntu__ [0] (
06:14:53 Part Nick_Brackley
06:15:23 Join Nick_Brackley [0] (i=3aa67da1@gateway/web/cgi-irc/
06:17:33 Quit Nibbier (Read error: 110 (Connection timed out))
06:17:51 Quit chrisjs169 (Read error: 110 (Connection timed out))
06:18:11 Join Nibbier [0] (
06:20:00 Quit rotator ()
06:21:57 Join gotthardt [0] (
06:23:59 Join jhulst [0] (
06:28:48 Part DanHibiki
06:35:01 Join ptw419 [0] (
06:40:27*amiconn wonders about the latest gigabeat commit
06:42:01amiconnI think this adaptive bass without no unit is about as bad as the behind-the-scenes volume capping we got rid of long ago
06:42:29gotthardti couldnt find a unit
06:42:38gotthardtbut it makes my bass sound better
06:44:48*jhMikeS wonders about the commit too and hopes it's not on by default.
06:45:10jhMikeSargh, it is
06:45:12gotthardtit is on by default - try and see if you like it
06:45:50gotthardteasy to turn off -
06:46:03Lloreangotthardt: We got rid of behind the scenes bass or volume reduction to prevent clipping. What exactly does "adaptive bass" do to prevent clipping?
06:46:22jhMikeSHaving the correct dB in the list partly helped. The volume limiting in sound.c requires correct values too.
06:46:32gotthardtits in the datasheet for the 8751 as the recommend way to prevent bass clipping
06:47:22 Join midkay_ [0] (
06:48:00gotthardtand i was getting bass clipping - esp during playback in car - thats why i put in the adaptive bass setting
06:48:16Lloreangotthardt: That doesn't answer my question, though.
06:48:29jhMikeSthat's true but I think proper volume tweaking would be better. I don't know why tone controls don't just use a precut/preboost/
06:49:08gotthardtLlorean: I can give exact answer - the datasheet does not give specifics - i track it down...
06:49:53amiconnThe general consensus is to give the user full control about the parameters. If that means clipping, let it clip, rather than changing settings behind the scenes
06:49:54Lloreangotthardt: Basically, what conditions did you get bass clipping under?
06:50:45gotthardtthe adaptive bass has control settings from 0 (off) to 15 max
06:51:24jhMikeSsomething went odd at some point because it seemed like clipping was gone for a bit and then came back
06:51:40amiconnAnd I wonder how one can get bass clipping in car... the only proper setting for hooking up a dap to a car stereo is line level with bass == treble == 0dB and eq and similar sound distrotion disabled
06:53:41BHSPitMonkeyanybody heard of Insignia mp3 players?
06:53:42*jhMikeS has a solution to the FIFO emptying on sansa but it limits available frequencies to divisions of the PLL :\
06:53:53BHSPitMonkeythey're pretty neat looking gizmos
06:54:00LloreanAnd if we're getting clipping with a level of 0dB then there's something else wrong instead.
06:54:20gotthardtLlorean: no clipping at 0 for me
06:54:32Lloreangotthardt: I'm talking about without the feature
06:54:38amiconnjhMikeS: Huh? I thought the fifo emptying problem is due to cache flushes being too slow at lower clock?
06:54:46Lloreangotthardt: As I asked, and you haven't answered, under what conditions were you getting clipping before?
06:54:54jhMikeSLlorean: like I was completely gona after setting the dB values correctly in the tone controls. I couldn't get clipping with all of them cranked up.
06:55:33amiconnThe clock setup doesn't disable fiq....
06:55:33 Quit Nick_Brackley ("CGI:IRC (EOF)")
06:55:34gotthardti get it in the speakers in my car with the bass on 4.5db
06:55:37 Join Nick_Brackley [0] (i=3aa67da1@gateway/web/cgi-irc/
06:55:38LloreanjhMikeS: Depending on the track, though, clipping should be expected as *possible* at any point where Volume + Bass or Treble or EQ filter is greater than 0, right?
06:55:40jhMikeSamiconn: it is...but we leave the pll set and switch clock sources from PLL/divider to PLL when flushing and back again
06:55:53Lloreangotthardt: If the bass is at 4.5, and the volume is at 0, clipping is to be expected, of course.
06:55:57gotthardtit wasnt on all tracks
06:56:02BHSPitMonkeyLol. Cheapo PDF user manual. It's literally a html site, viewed in IE, and printed to a PDF printer.
06:56:10BHSPitMonkeyYou can see their local filename as the URL at the bottom.
06:56:15amiconnEH, that sounds nasty
06:57:03jhMikeSLlorean: true, but EQ has manual precut and clipping shouldn't be possible if a volume reduction exactly compensates the increase from the control.
06:57:33LloreanjhMikeS: Yes, doesn't that go along with what I say? If you set a precut equal to the increase, the final level is still 0?
06:57:47LloreanjhMikeS: I'm missing something here, I think.
06:59:31jhMikeSLlorean: I think I was. As long as a PCM sample of 0db = 0db as output voltage we're fine. yes.
07:02:02LloreanjhMikeS: Yes, and that was my point. If it doesn't, there's a real problem, and so whatever this "adaptive bass" is probably not the solution. If it does equal 0dB of output voltage, then we don't need it in the first place. Right?
07:02:05 Quit midkay (Read error: 110 (Connection timed out))
07:02:36LloreanThough I don't see a problem with adding it as an option in the menu or something, for people who want it, but it doesn't seem a good 'default' to me.
07:03:39***Saving seen data "./dancer.seen"
07:04:32gotthardthere is from the datasheet:
07:04:34gotthardtAdaptive bass boost: The bass volume is amplified by a variable gain. When the bass volume is low, it is boosted more than when the bass volume is high. This method is recommended because it prevents clipping, and usually sounds more pleasant to the human ear.
07:05:27*jhMikeS is updating to have a listen
07:05:38gotthardtThe chip does the work with 15 settings - 0 is off
07:06:41amiconnThat sounds like it performs bass dynamic compression - a thing that should never be the default imo (or rather, like in this case, the only way to use the setting)
07:06:46LloreanSo, rather than applying say +4.5 to all bass, if it's a quiet portion of the song it gets more, and a louder portion gets less? This definitely sounds like something that shouldn't be on by default or lock on.
07:06:51jhMikeSI was the one that moved the driver over so I know what it does. I just never even tried the adaptive setting.
07:07:35gotthardtThe default setting is off
07:07:58gotthardtSo it should sound the same for linear or adaptive at the off setting
07:08:59amiconnYes, but if I had a gigabeat, I couldn't set bass to e.g. well defined +6dB now
07:09:08amiconn...which is bad, imo
07:09:30gotthardttrue - you would have to comment out the define in the gigabeat config
07:09:33amiconn<unrelated>Running PP with disabled cache is funny</unrelated>
07:09:37gotthardtand then rebuild
07:09:54*jhMikeS still wants to know how clipping came back after the problem disappearing for awhile
07:11:47jhMikeSamiconn: you think it's nasty to switch clocks? the tracing in the emu shows this sort of use.
07:12:28amiconnI think it's nasty to limit our ability to switch clocks
07:12:46amiconnI'd rather want partial cache flushes
07:13:14Lloreangotthardt: And no user should have to rebuild to be able to actually use the full range of bass we supposedly "offer"
07:13:51gotthardtok - so how can we offer all that a codec can do?
07:14:21gotthardtif the manufacturer recommends a different setting, should I make a new menu to select linear or adaptive?
07:14:40jhMikeSamiconm: me too but I'm not seeing evidence of it being done on sansa
07:15:17Lloreangotthardt: I don't see why not, so long as the "Bass" menu is updated to represent realistically what is being offered when in adaptive mode.
07:16:18jhMikeSthere's no bass cut in this mode either
07:17:38gotthardtsounds like I should just comment out the define and recommit
07:19:02jhMikeSthere's some details like not having sound.c compensate the volume when using that mode for the bass (but it still should for treble)
07:20:17 Join EbErT [0] (
07:21:05jhMikeSWhich compensation I believe was added for WM codecs because others considered the facility on chip insufficient iirc
07:21:55jhMikeSOn the bright side, the available level of boost seems greater using that.
07:23:20 Join tedr0ck [0] (
07:24:38scorcheheh...if only it was this easy:
07:25:41*amiconn now really wonders what 0x70000034 does
07:26:00*jhMikeS apolized for his kneejerk whining about the change and finds it to be quite of benefit afteral :)
07:26:26amiconnIt's definitely not ram timing. Changing the values and then running speed test from dram with cache disabled doesn't change anything
07:26:59amiconnAll I know now is that disabling the cache makes code run ~7 times slower from dram
07:27:32LloreanjhMikeS: It makes the "Bass" setting misleading, and removes user choice for what Bass actually ends up at.
07:27:43LloreanEven if it sounds better, it needs to both be clear what's going on, and probably be off by default.
07:28:23jhMikeSLlorean: I agree (and I myself tend to hate options) that the user should have a mode choice with this.
07:29:33jhMikeSI think all the WM codecs with tone contols can use it.
07:31:12markungotthardt: hi!
07:31:22gotthardtmarkun: hi
07:31:29LloreanjhMikeS: Does that include the one in the 5G where the tone controls don't seem to do anything right now but increase volume?
07:31:41jhMikeSamiconn: that's why I fear a non-cacheable framebuffer btw
07:32:00amiconnYou don't run code from the framebuffer....
07:32:00jhMikeSLloean: hmmm...wasn't aware of those problems
07:32:31 Nick midkay_ is now known as midkay (
07:32:35amiconnLlorean: The G5 has a different chip that doesn't have the standard tone controls
07:32:48jhMikeSamiconn: the cached dram for data is also a good deal slower too
07:32:56amiconnIt has the registers for hw eq, but with just 2 of the 5 filters actually working
07:33:05amiconnYes, sure
07:33:13Lloreanamiconn: We should really either disable them, or replace them with the software tone controls until someone figures out if there's useful ones.
07:33:35amiconnBut there shouldn't be much difference whether you write the data to uncached ram, or write it cached then flush long as you don't rewrite multiple times between lcd updates
07:34:52amiconnLlorean: I will change the tone controls on G5 to use the eq low shelf and high shelf (because that's what they actually are)
07:35:01jhMikeSbut perhaps the burst when it flushes to ram during the write is much faster that writing word by word
07:35:09amiconnThis means we have true hw tone controls, but no hw eq
07:35:36Lloreanamiconn: Works for me. As long as the menu settings actually do something that seems like Bass/Treble, I'm happy. :)
07:35:58amiconnOf course I will also keep the ability to select cutoff frequencies (which does work)
07:36:13amiconnThere will be 'Bass cutoff' and 'Treble cutoff'
07:36:30amiconnWe could also offer this for UDA1380 targets
07:37:18amiconn...which has 2 cutoff frequencies for treble/bass.
07:37:21jhMikeSWm8751 has optional cutoff too. one thing with hardware contols is they are samplerate dependent but I'm not sure it matters here.
07:37:26amiconnUnfortunately one bass cutoff allows 24dB boost, the other only allows 18dB. I'm not sure how to handle this
07:38:45amiconnI tend to think that nobody uses >18dB (because it sounds like our famous bag of shit) and we could limit the setting to <=18dB
07:39:01amiconnBut there are weird users out there... ;)
07:39:57jhMikeShehe...I can think of some neighbors that would probabaly want +100dB :p
07:40:36LloreanLike those people who set bass to max, then software EQ it up further, and talk about how great Rockbox sounds because of how much it lets you mangle the sound?
07:40:53 Quit tedrock (Read error: 110 (Connection timed out))
07:40:59amiconnjhMikeS: How many different settings are there? The WM8758 has 4 each for treble/bass
07:41:09jhMikeSjust for the subwoofer in their trunk so they can blow the panels off the car
07:41:16imbecilehow do i manually install themes? ive been googling and havent found the answer yet
07:41:42imbecilein an 5g ipod
07:41:51amiconnjhMikeS: That can be done easier and more reliable with a bit of explosives... ;)
07:41:59jhMikeSone cutoff option for treble, one for bass
07:42:11*jhMikeS likes amiconn's thinking
07:42:35Lloreanimbecile: Assuming they're created properly, you just extract them directly to the player. Otherwise you'll need to look through and put files in their appropriate directories. Look at a working theme .zip to see an example of where each file goes.
07:43:36amiconnjhMikeS: 'one setting' meaning what?
07:44:29amiconnWM8758 allows 80, 105, 135 and 175Hz cutoff frequency for bass, and 5.3, 6.9, 9 and 11.7kHz for treble
07:47:27jhMikeSBass: 130/200 at 48kHz, Treb: 4/8KHz at 48KHz
07:48:42amiconnUDA1380 allows 250/300Hz for bass, and 1.5/3kHz for treble (at 44.1kHz)
07:49:04amiconnThese values don't look right to me...
07:49:31jhMikeSI guess that mean at 44100kHz its 119.4375/183.75 and 3675/7350
07:50:08amiconnUDA1380 has only one setting that switches both treble & bass cutoff
07:50:37*amiconn should check those values by measuring
07:50:37jhMikeSI do seem to remeber they're sample dependent there too
07:55:47 Join stripwax [0] (
07:57:24jhMikeSclock switching isn't _nescessary_ though. changing the FIFO tresholds also causes an FIQ so it fills up. they can then be reset after the flush.
08:05:08jhMikeSamiconn: another register of interest might be 0x600060a4. the sansa OF never seems to alter 0x600060a0 either afaik (which set_cpu_frequency currently does).
08:10:08 Join kfazz [0] (
08:11:10markunLlorean: I want to post a link in the forum to this 'player'
08:11:56amiconnjhMikeS: 0x600060a0 only seems important on the ipod color/photo
08:12:02Lloreanmarkun: Under new ports?
08:12:18markunnot sure if I should hijack any of the existing threads in New Players (like the N3 thread or 'oposite of porting'
08:12:21amiconnBits 30 and 31 must be zero during clock changes or it's highly unstable
08:12:22markunor just start a new thread
08:12:36markuneh yes, new ports
08:12:48LloreanI'd say start a new one
08:12:56amiconnThe color/photo is the only target where these bits are set, but the clock change routine does this zeroing on all PP502x ipods
08:13:06amiconnDoes no harm if the bits are already zero
08:13:31amiconnWhat does the sansa do to 0x600060a4
08:13:46Lloreanmarkun: Do you intend to contact them, since Rockbox might even be a good way to get firmware for the device off the ground a good deal more quickly?
08:13:47amiconnI still didn't manage to find the clock change routine in the sansa OF
08:14:51markunLlorean: I could, didn't think about that
08:15:49LloreanIt'd probably make a lot of sense for them too, since it provides a good starting point, and means a lot less work on application style stuff.
08:16:27jhMikeSamiconn: it does some manipulation with it but it ends of setting it to 0x23
08:17:00jhMikeSIt's set to that in i2c-pp.c and seems to affect the speed of that but I'm not sure what else.
08:17:25jhMikeSI put it there since it helped alot
08:18:30markunLlorean: looks like USB is full-speed only :(
08:18:59markunbut you can use a SD/MMC adapter of course
08:19:33LloreanWell without internal storage, you probably want to anyway
08:19:54ptw419through the gigabeat-s branch are we going to have to send commits through patch tracker?
08:19:58LloreanIt's clearly a video device (well, video and gaming), and with that end in mind, a single 4gb card is probably not enough
08:20:07markunptw419: no
08:21:08ptw419do we get commit access to the gigabeat-s branch?
08:21:21ptw419or can i if possible
08:21:38ptw419or tell me how to do it if its done some other way ;)
08:22:40 Quit stripwax_ (Read error: 110 (Connection timed out))
08:22:49markunif the branch has been setup just for you and aliask I don't see why you wouldn't get commit access
08:23:17ptw419yes, problem was zagor set it up, but never gave us a username and p/w :(
08:23:43scorchewell, he should be up in a few hours...
08:26:30aliaskUnfortunately I have work in an hour.
08:29:21markunLlorean: ok, made a thread
08:29:54GodEater_it's alerady 8:20 am in Sweden - should be up in less than a few hours I'd have thought
08:30:16Nick_Brackleyaliask: i posted about your superdom patch in flyspray again
08:30:26aliaskI saw, but I've been a little busy
08:30:34aliaskI can take a look in a minute if you liek
08:30:35Nick_Brackleyno worries
08:30:55Nick_Brackleyaliask: yeah if you have the time of course
08:32:10scorcheGodEater: an hour from now is the typical time, i suppose
08:36:23GodEater_lazy people - I'm already at work!!!
08:37:04 Join ender` [0] (
08:37:08Nick_BrackleyGodEater_: nicely done ;)
08:37:40GodEater_for being at work or some other random thing I've done ? =/
08:37:56Nick_BrackleyGodEater_: the work thing
08:38:01GodEater_ah - thank you :)
08:47:55 Join Rob2222 [0] (
08:51:33 Quit maffe (Remote closed the connection)
08:52:08aliaskAh markun you saw that hackaday post too - looks interesting doesn't it.
08:52:28*scorche lubs hackaday
08:54:28Lloreanmarkun: I modified your post. If I understand correctly, the relevant hardware is mostly the "VCMX212" board, right?
08:54:35LloreanWell, I modified the title of your post, rather
08:56:39aliaskNick_Brackley: I added a new patch to the game, it should work fine now.
08:57:25 Join dan_a [0] (n=dan_a@
08:57:44Nick_Brackleyaliask: thank you, will test it later and post back on flyspray
08:58:40 Quit JdGordon ("Konversation terminated!")
09:01:10markunLlorean: ok, no problem
09:02:07 Part Llorean
09:03:04 Join Zagor [0] (n=bjorn@rockbox/developer/Zagor)
09:03:41***Saving seen data "./dancer.seen"
09:04:23*GodEater_ waits for ptw419 and aliask to pile into Zagor
09:05:17aliaskAh, thanks for the ping GodEater_ :)
09:05:30aliaskZagor: Would it be possible to get write access to SVN?
09:05:40aliaskOr at least the gigabeat s branch.
09:06:13 Quit Rob222241 (Read error: 110 (Connection timed out))
09:06:28 Join webguest08 [0] (i=cfbfd7d3@gateway/web/cgi-irc/
09:06:39Zagoryeah, that's the plan. our current version of svnserve doesn't support path-based access so you'll have to make sure yourselves that you're only working in the branch.
09:07:07GodEater_with great power comes great responsibility etc. etc.
09:07:22 Join Paprica [0] (n=Paprica@
09:07:27Zagorsomething like that :-)
09:07:42 Join petur [0] (n=petur@rockbox/developer/petur)
09:07:45aliaskIndeed. Well, I suppose that's the great thing about version control. I could majorly screw up and you could just revert it :)
09:08:11aliaskBut I'll double check everything. Perhaps I should learn how to use svn to commit before trying anything...
09:08:25 Quit markun (Remote closed the connection)
09:09:02*GodEater_ looks forward to the juicy content that is sure to be in the latest "NNNNEEEEEEEDDDD HEEELLLPPP" thread to appear on the forums
09:10:07scorcheGodEater: you enjoy that one post in the nano 2g thread? =P
09:10:43ptw419scorche : I wish it would have been that easy for the S :)
09:11:11Zagoraliask, ptw419: email me the login and password you want:
09:11:21 Join markun [0] (n=markun@rockbox/developer/markun)
09:11:23ptw419ok, will do
09:11:23GodEater_scorche: I just didn't even know where to start with that guy ;)
09:11:38scorcheptw419: and all we need to do to get music is do play_music(helloworld.mp3_
09:11:45ptw419scorche: I did email Badger, I'm not sure if/when he's around
09:11:46scorchewell, minus the typos
09:12:14scorcheptw419: yeah..i need to speak to him too, but i didnt see him before i went to bed and missed him by the time i got home
09:12:20 Quit Nick_Brackley ("CGI:IRC (EOF)")
09:12:24 Quit webguest08 ("CGI:IRC (EOF)")
09:12:30 Join Nick_Brackley_1 [0] (i=3aa67da1@gateway/web/cgi-irc/
09:12:58scorchehrm...i guess i shall respond tot he guy
09:13:08aliaskZagor: Done.
09:14:19 Join Llorean [0] (n=llorean@
09:14:39GodEater_scorche: which one ?
09:14:57scorcheGodEater: nano 2g with a concise and simple answer
09:15:14GodEater_scorche: something like "your oversimplying so much I just don't even know where to start?" ;)
09:15:16 Quit Seed (Nick collision from services.)
09:15:18 Join austriancoder [0] (n=austrian@
09:15:21*GodEater_ puts his "Jerk" cap on
09:15:24 Join Seed [0] (
09:15:34scorchenot really, but now that you mention it, sure! =P
09:15:34 Part EbErT
09:15:51 Join Guile` [0] (n=Guile@
09:17:00ptw419aliask i think there may only be two framebuffers, one at 0x80100000, the other at 0x84100000
09:17:14aliaskWhat are the odds! :P
09:17:17ptw419at least w/ those offsets of 0x100000
09:17:49GodEater_ptw419, aliask : is there still a supplier for new Gigabeat S's ?
09:17:55GodEater_or am I looking at Ebay again ?
09:18:08aliaskGodEater_: I'm pretty sure they are still sold in retail
09:18:12ptw419i never knew there was a supplier
09:18:15aliaskAt least, they are in Australia
09:18:35ptw419yes, or maybe on ebay too
09:18:37GodEater_what do they go for down there? specifically the 60
09:18:56*stripwax can't get rbutil to compile, grr
09:18:58ptw419i've seen the 30's for around 150, sometimes there are good deals for like 100
09:19:06ptw419but thats on ebay
09:19:12GodEater_stripwax: linux ?
09:19:14aliaskThat's brand new
09:19:22*GodEater_ goes to check the exchange rate
09:19:33*dan_a does the same
09:20:05GodEater_not too bad
09:20:07Zagoraliask, ptw419: do we have your real names somewhere? entering it in your irc clients would be a welcome start imho.
09:20:24aliaskWill Robertson
09:20:35ptw419mine is on the wiki
09:20:43ptw419James Espinoza
09:20:47GodEater_resisting urge to make "Lost in Space" gags
09:21:00aliaskTHAT'S WILL ROBINSON
09:21:10scorcheclose enough
09:21:13Battousaimust get that a lot
09:21:16stripwaxGodEater - windows/cygwin , unfortunately. but wxwindows built ok so that's promising..
09:21:18GodEater_I knew you'd say that
09:21:42*dan_a is tempted at the idea of an S30 at 150AUD
09:21:43aliaskZagor: Where abouts should I put it in my IRC? Description?
09:21:44GodEater_stripwax: shame - builds flawlessly on gentoo :)
09:21:58GodEater_am I right in thinking the S30 lacks the FM receiver ?
09:22:11scorcheGodEater_: you should send a box to him and try not to write will robinson instead =P
09:22:20ptw419S30 has fm reciever
09:22:29GodEater_scorche: I trying *not* to make the gags remember ;)
09:22:41aliaskReally? According to the wiki it doesn't. It should be updated...
09:22:48GodEater_yeah - that's what made me ask
09:22:54 Quit ^BeN^ (Read error: 110 (Connection timed out))
09:23:34scorcheGodEater: i was trying to find something amusing to put inside the box (just as the robot), but all i could find around was a mini crunch bar and a free toilet leak tester tablet (this is a true story btw)
09:24:11*Llorean needs to look for an S## Gigabeat too now, out of some sick addiction.
09:24:20GodEater_a toilet leak tester tablet
09:24:24scorcheLlorean: you cant handle the power!
09:24:27GodEater_my world view is changed again
09:24:43 Join miepchen^schlaf [0] (
09:24:55scorcheGodEater: i found a dispenser at home depot
09:24:58GodEater_Llorean: we only want them so that amiconn can shout at us and go "what's the point? It's too powerful!"
09:25:27scorcheand took one while wondering what to do with it...luckily, the gigabeat S had arrived and i put it in there before sending to little mister will there =P
09:25:46stripwaxgah, rbutil build error, in streambuf.tcc in cygwin includes..
09:26:06Zagoraliask: I thought all clients had a "real name" field. Maybe yours is called "description"? It's the field that currently says "New Now Know How".
09:26:30ptw419let me enter mine real quick
09:26:32 Quit ptw419 ()
09:26:54 Join ptw419 [0] (
09:27:12stripwaxabout one third of my build error here - - anyone have an idea about that? (win32 sim builds fine on my cygwin installation btw)
09:27:14 Join desowin [0] (
09:27:16 Quit aliask ("ChatZilla [Firefox]")
09:27:26 Join aliask [0] (
09:27:42Zagorvery nice
09:28:08GodEater_scorche: nice post - nearly word for word ;)
09:28:48GodEater_could you put maybe a "(tm) GodEater" ? ;)
09:29:21scorchei didnt hear any restrictions when you said it
09:29:27Zagorok guys, you're added. simply check out svn:// and you're set
09:29:47aliaskI've already checked out - does that matter?
09:29:49GodEater_scorche: no no - you're right - too late now
09:29:50 Quit stripwax ("Miranda IM! Smaller, Faster, Easier.")
09:30:07Zagoraliask: no, that's fine. it will ask for your user/pass when you commit
09:30:17markunZagor: how does a branch work? are all the commits from the main tree also committed to it?
09:30:26aliaskGreat. Now I gotta make this thing compile so I can commit and get to work.
09:30:31Zagormarkun: no, it's completely separate
09:30:40markunthen what's the advantage?
09:30:41ptw419ty Zagor
09:31:02GodEater_markun: the advantage is they can't break the main trunk
09:31:19Zagorand having it on our server makes it visible for all
09:31:30GodEater_markun: with the caveat that later it will be a bit more work to merge it back in
09:31:59markunwill it keep all the commit logs after the merge?
09:32:07GodEater_markun: should do
09:32:25Zagorthe gigabeat boys should merge from the main trunk regularly, so it doesn't pile up to a huge diff
09:32:35markunI still think git would be very nice in this case :)
09:32:51GodEater_markun: nothing stopping them using git-svn should they choose
09:32:55 Quit Guile`` (Read error: 110 (Connection timed out))
09:32:55Zagormarkun: how would git make any difference in this case?
09:33:14markunZagor: they just checking the changes from main rockbox and from eachother
09:33:21markunor can you do that with a branch as well
09:33:35markunmaybe I should start reading before I talk :)
09:33:57GodEater_dan_a is Mr. Git fan extraordinairre as I recall - maybe ask him
09:34:20GodEater_in fact I expect him to read the wiki page I did and add corrections ;)
09:34:24markunI just remember we had a hard time merging with our changes back in the rockbox tree while working on the gigabeat f port
09:36:03GodEater_I agree it'd be nice if they did use it - I'd like to pull from them too - but I wouldn't force anyone to use it
09:36:57GodEater_having said that though - getting a git repo set up locally that others can pull from was non-trivial :(
09:37:11GodEater_all the guides out there assume you'll use someone ELSE to do that
09:37:28ptw419I'm open to any ideas
09:37:39aliaskCan I omit the filename in svn commit to get it to commit all changes in the tree?
09:37:50ptw419but atm we need a way to get all our work put together
09:37:59ptw419so we know what we hae
09:38:00scorcheGodEater_: lostlogi is a fan of it as well =)
09:38:02Zagoraliask: yes. it works recursively from where you are.
09:38:02GodEater_ptw419: I think leave it for the time being then ;)
09:38:23ptw419GodEater: I agree, for the time being
09:38:28GodEater_if you guys are only just getting to grips with svn, then chucking another tool into the mix you've never used before is not sensible
09:39:06GodEater_scorche: well then why hasn't he added more stuff to the wiki page then ? ;)
09:40:00scorchehe is a busy man
09:41:29aliaskTata all
09:41:31 Quit aliask ("ChatZilla [Firefox]")
09:43:30ZagorFYI: commits in the branch will trigger commit mails, but they will not be shown on the front page
09:43:51GodEater_will they trigger builds ?
09:44:04ZagorI hope not :-) I'll check.
09:45:15Zagordoesn't look like it. will just committed a bunch of things, and no servers are building.
09:45:35GodEater_isn't it a 60 second check ?
09:45:52Zagorah, right. let's wait and see then. :-)
09:46:18ZagorI'm not sure where Bagder keeps the master script.
09:46:46GodEater_up his sleeve somewhere I think
09:47:03scorcheZagor: speaking of, do you know if he will be around soon?
09:47:05Zagorbut most likely it's doing a 'svn log' against the trunk, which won't see commits in the branch
09:47:16GodEater_good :)
09:47:38markunthe forums have become terribly slow here, does anyone notice the same?
09:47:48GodEater_markun: I always get that this time of morning
09:47:53scorchemarkun: they typically are around this time
09:47:57Zagorscorche: I don't know. He's on vacation, but I don't know when he'll be home/at his computer.
09:48:09scorcheZagor: oh?...i thought he was back by now
09:48:15scorchewhen does he get back?
09:48:29scorcheor was that the "i dont know" part?>
09:48:39Zagori'll see if I can find out
09:48:53 Quit dan_a ("using sirc version 2.211+KSIRC/1.3.12")
09:49:56Zagorscorche: looks like he'll have vacation until the 29th.
09:50:17 Join printfXh4 [0] (
09:50:17Zagorthe "i don't know" part is I don't know how much he'll be away and how much he'll be home and possible on irc
09:50:29scorcheare you able to discuss with me on copyright technicalities? =P
09:51:07 Join MonkeyTamer [0] (
09:52:30scorchewell, we were putting together the final version of the guidelines for the new theme repository, and we could use a definitive stand to take on the issue of copywrited DCW, we decided that "inspired by is fine as long as images are long as it is not identical which it wont be with our WPS system", however we werent that versed on the entirety of the issue...and i suppose copyrighted images is a "just say no" thing
10:02:29 Join Thundercloud [0] (
10:05:44 Join maffe [0] (
10:06:41 Join davina [0] (
10:09:37*scorche inserts a Zagor: into the above
10:11:17ZagorI agree. copyrighted material should be a definite no-no.
10:11:40scorcheand the inspired by portion?
10:12:16Zagorinspired by is fine. I think standard copyright law is a good guide here.
10:12:17 Part maffe
10:12:40 Join maffe [0] (
10:12:41scorcheis there any way to bring fair use into the equation with copyrighted images?
10:12:52LloreanZagor: We just weren't sure where standard copyright law would land on say, an interface mimicing say, the exact layout of another interface but with brand new graphics.
10:13:01LloreanOr near exact layout, even.
10:13:04Nick_Brackley_1Zagor, scorche: a quick question, what are the rules on copyright used with permission?
10:13:34ZagorNick_Brackley_1: generally no, unless "permission" is something like a CC license
10:13:56Nick_Brackley_1ok thanks
10:14:03Zagorotherwise permission is difficult to verify and convey to the future persons
10:14:55Nick_Brackley_1yeah makes sense
10:15:11LloreanZagor: The current guidelines basically say "You're licensing all of your work under CC-SA3.0, any included parts and if any part is copyright someone else, include the license itself that it falls under"
10:15:28ZagorLlorean: interface copyright is not part of international copyright law, it's strictly a US thing. I think we shouldn't worry about it until it becomes an issue.
10:15:51 Quit Lynx_ (" HydraIRC -> <- State of the art IRC")
10:16:02scorcheZagor: and is that a no to the fair use question (i wasnt sure where the grounds on that were at all)
10:16:03LloreanZagor: Okay. I know in the past there's been a lean toward using US guidelines, but I'm happy with a "we'll assume there's elbow room there unless a problem happens"
10:16:41LloreanI'm *almost* certain I wrote something like "If anyone complains, we'll take it down and let you know what needs to be cleared up before it can be reposted"
10:17:05scorcheLlorean: well, if you didnt, i will ad dit during my formatting
10:17:19Zagorscorche: fair use is a very gray area. I would prefer if we avoid relying on it.
10:17:30Lloreanscorche: Though you might want to add "If anyone complains, and there is any reasonable ground to their claim"
10:18:17LloreanZagor: That's pretty much my standpoint on fair use. It's hard as hell to make it stick when you need it, and it's gray as to what it even covers unless you're educational (and even then it's hard to rely on)
10:18:33ZagorLlorean: actually we've been pretty consistently turning our backs at the US legal problems with patents and so forth, and referring to swedish law whenever actual laws are cited.
10:18:57LloreanZagor: Hm, I seem to remember some statements regarding US for copyright. I'm probably wrong.
10:19:19Bagder(for the record, I'm on and off here during my vacation that lasts until the end of July)
10:19:23Zagoryeah we've probably discussed it.
10:19:43scorcheBagder: long vacation =) ...having fun?
10:20:06LloreanOne important question: Where's the hosting of the themes site. The question of interface copyright can probably simply be decided by whether they're hosted in the US, or "internationally"
10:20:13Bagderus swedes are spoiled with long vacations, and yes I'm having a good time!
10:20:30 Join obo [0] (
10:20:50ZagorLlorean: I was going to ask the same thing :-)
10:21:07scorcheLlorean: 1&1, which is hosted in a few places...likely the US
10:21:45Zagoractually the server looks to be in germany
10:22:00*scorche references the "a few places" remark
10:22:31Zagorhow is it a few places? is a fixed IP address. or do you mean it links to other places?
10:22:52scorcheZagor: i am referringt o the whole of 1&1...i hadnt gotten to the actual IP yet
10:23:34scorcheand yeah, it does seem to be germany
10:23:35ZagorBrauerstrasse 48, Karlsruhe, Germany :-)
10:23:50Zagorwell, at the last owner of the ip block
10:23:50scorchephone: +49 721 91374 50
10:32:16 Join webguest76 [0] (i=c023110a@gateway/web/cgi-irc/
10:33:08GodEater_completely OT - but I shall be flying the Rockbox flag here tomorrow :
10:33:51scorche...those crazy british...
10:35:22GodEater_I say sir, we're not crazy - we're eccentric.
10:35:36scorcheyou are only eccentric if you are rich
10:35:48austriancoderpetur: ping
10:35:48GodEater_how about moderately well off ?
10:35:56scorchenope...just rich
10:35:57peturaustriancoder: pong
10:36:17GodEater_tell me - am I crazy for going if there's free booze involved ?
10:37:10austriancoderpetur: I check the endpoint caps field and get this:
10:37:11austriancoder 0x0x128080 -> 100101000000010000000 active IOC LEN: 10010 = 18
10:37:16 Join nekura42 [0] (
10:37:19austriancoderlooks perfectly
10:37:35scorcheno, wish something like that happened around here...that is utterly "bonkers"!
10:37:58GodEater_scorche: it only happens when people organise it - so get organising ;)
10:38:05austriancoderpetur: the linked list seems also to be okay.. as trail and head are pointing to 0x40008C00
10:38:41peturaustriancoder: where does "arcotg_ep_queue, bad params" come from?
10:39:41austriancoderpetur: I am waiting for an ENDPOINTCOMPLETE register change.. so that i can kick the request from the queue... !list_empty(&req->queue)
10:40:06austriancoderno.. that i can kick the transfer description from the request
10:40:19GodEater_I need a costume. I think I need a pipe and some braces. Perhaps a monocole.
10:40:35austriancoderpetur: if you want I can uncomment it and see what happens..
10:40:46nekura42and a cane while you're at it :)
10:40:48scorchei am sure i am thinking of the wrong kind of braces, here
10:41:11GodEater_I think you call them "suspenders" in the US
10:41:15GodEater_things to hold your trousers up with
10:41:33 Join safetydan [0] (n=dan@rockbox/developer/safetydan)
10:41:37GodEater_suspenders are lingerie only in the UK
10:41:43austriancoderpetur: please fetch
10:42:09austriancoderpetur: and i can tell you where I am not 100% sure if code is correct
10:42:27peturaustriancoder: problem is 1) I'm at work now so 2) it would be nice to have debugging sessions in the evening and 3) I have no player myself to to try the code - but I'll have a look
10:43:11austriancoderpetur: no problem.. there are enought other talentet coders here :)
10:43:31*austriancoder seeks for a talented coder with a Sansa
10:45:12nekura42I would like to perhaps someday collaborate with someone on USB functionality on the iPod
10:45:47peturnekura42: meet austriancoder
10:45:58peturaustriancoder: meet nekura42
10:46:37austriancodernekura42: fetch current patch..
10:46:42LloreanWhy does my "svn update" say I'm at revision 13869 if the last revision is 13867?
10:47:20peturwould the S branch have anything to do with it?
10:47:29scorcheaustriancoder: i believe she said ipod..not sansa =O
10:47:34nekura42I'm from the #ipodlinux kernel, but I imagine we can share hardware knowledge and what not when it comes to getting USB up and running itself
10:47:41 Part maffe
10:47:47austriancoderscorche: portalplayer is portalplayer
10:47:48Lloreanscorche: Sansa ought to be semi-relevant though, no?
10:48:01Lloreanpetur: Hm, an interesting thought.
10:48:14scorcheaustriancoder: i thought "otg" was somewhere in there....heh
10:48:49austriancodernekura42: ah okay... rx of requests works.. but I dont know why tx fails at the moment
10:49:18*austriancoder thinks that his code should run on ipods too
10:49:31ptw419petur: that may explain it b/c there have been only two commits to the S branch so far
10:50:01*nekura42 nods
10:50:02 Join maffe [0] (
10:50:12LloreanI couldn't pull up logs for 13868 and 9.
10:50:20 Quit maffe (Remote closed the connection)
10:50:51ptw419Revision 13869 in gigabeat-s branch
10:50:55 Quit TiMiD[FD] ("leaving")
10:50:57LloreanAlright then
10:50:59LloreanConfusion solved
10:51:03 Join pixelma [0] (i=pixelma@rockbox/staff/pixelma)
10:52:03austriancoderanyone wants to try my usb stuff on an ipod?
10:53:18midkayaustriancoder: what usb stuff? :)
10:53:52austriancodermidkay: my current work of soc
10:55:05Nick_Brackley_1is this usb-otg?
10:55:24LloreanNick_Brackley_1: It's likely to contain a lot of the work necessary to begin USB-OTG support.
10:55:39midkayaustriancoder: so , "usb stack".. is it a rockbox disk mode for ipods?
10:55:47midkaywell, not just for ipods, but..
10:56:04Lloreanmidkay: Sansas, but it should be quite useful for iPods too.
10:56:27peturat the moment it is only USB DEVICE, the OTG part is still a dream ;)
10:56:51austriancodermidkay: it will be .. at the end of this summer... at the moment I have a problem with sending an answer back to the host ... and here I need some help
10:56:58Nick_Brackley_1i have an iriver h300, and am willing to test anything you guys can throw at it ;)
10:57:12midkayaustriancoder: cool! well sure, i'd be glad to test. i have an ipod 5G.
10:57:46austriancodermidkay: you need to build it with LOGF support
10:57:50midkayaustriancoder: oh dear.. i'm not set up to compile at the moment :(
10:58:19austriancodermidkay: okay.. will compile it for you
10:58:27midkayif you can build one and compile it, yes, that'd be great.
10:58:33midkay(60gb thus 64MB ram)
11:00:01austriancodermidkay: IPOD video or?
11:00:07midkayaustriancoder: yep.
11:02:15austriancodermidkay: 64 MB ram?
11:02:44*austriancoder is compiling
11:03:22 Quit webguest76 ("CGI:IRC (Ping timeout)")
11:03:27midkayaustriancoder: yep, sorry
11:03:42***Saving seen data "./dancer.seen"
11:04:14nekura42Ok, well, it's 5 AM now, so it's too late for me to start anything.. however, I saved a bookmark of those pages and will be downloading the Rockbox source at some point. I'll then apply the patch and take a look. My test bed will be a self-contained executable running on a 5.5G 30GB iPod Video under the iPodLinux kernel.
11:04:57nekura42I'll report back with results once I get that far.
11:06:24*austriancoder is uploading
11:08:56 Join bdgraue [0] (
11:09:44austriancodermidkay: 1) download precompiled rockbox and install it... 2) diconnect usb and start rockbox... 3) go into debug/developer/logf... 4) connect usb 5) after some time make a logf dump and put it online or mail me it - here is the link
11:10:35midkayaustriancoder: gotcha, installing
11:10:58 Join XavierGr [0] (
11:13:43midkayweird, some things are odd with this build, the last 3 root menu items are invisible and some messages seem broken, but i guess it worked.
11:14:11 Join spiorf [0] (
11:15:27midkayand what a lovely time for my host to go down :)
11:16:05midkayaustriancoder: email?
11:16:33 Part Llorean
11:16:58XavierGrgah austriancoder, you sure have guts to post your mail in a logged channel...
11:17:05*petur installs ubuntu in a virtualbox....
11:17:17midkayaustriancoder: sent.
11:17:30austriancoderXavierGr: I trust in gmails spam filter... but yeah.. maybe not so a good idea
11:18:44austriancodermidkay: thanks
11:18:55midkayaustriancoder: no problem, let me know if you need anything else :)
11:19:01austriancodercool... runs on ipod video too :)
11:19:28midkayseems to, except for those glitches i ran into.. maybe because it's a logf build?
11:19:33midkayvery bizarre stuff.
11:20:33austriancodermidkay: a friend of my has an Ipod.. so I will do some testing with his ipod, if he allows it
11:21:22midkayaustriancoder: cool :) hmm, how close do you think you are? have you just hit a small glitch or are you still in the early stages of coding?
11:22:22austriancodermidkay: I have some parts of a serial driver and looked a little bit in the storage driver.. at the moment I need to find out why tx fails
11:23:12midkayaustriancoder: sounds like you know what you're doing, which is always good :)
11:23:55austriancodermidkay: I have spend a lot of time in debuging this issue.. I hope I get it running today...
11:24:19 Join Llorean [0] (
11:24:23midkayaustriancoder: excellent, i'm looking forward to hearing about your progress.
11:26:38 Join webguest76 [0] (i=c023110a@gateway/web/cgi-irc/
11:37:06Lloreanlinuxstb: Copyright restraints. Either there need to be human gatekeepers or someone needs to go through and weed out any that seem iffy or in violation
11:37:46MonkeyTamerhmm, the change on the gigabeat for bass is slightly confusing, 0-15 now instead of the former decibel scale
11:37:46peturLlorean: Cassandra allready weeded quite a bit during devcon
11:38:09*scorche wonders where the question is that Llorean responded to
11:38:29Lloreanpetur: Yes, but the idea is that instead of requiring occasional weeding, with the new site (which is also more functionally useful anyway) we just have a human look 'em over before they get posted.
11:38:33Lloreanscorche: In the forums.
11:39:01Lloreanscorche: It's easier to discuss in channel, if he was around and had further questions.
11:39:09LloreanI'm also typing up a response there, for anyone else curious.
11:39:11scorcheLlorean: i think it would be best to reply in the thread though for others
11:39:40markunMonkeyTamer: ask gotthardt
11:39:46Lloreanscorche: Did I mention "junk files, like the .db file" in the guidelines?
11:40:17LloreanIsn't someone going to revert Goddhardt's change until it can be put in the menu properly?
11:40:23LloreanGotthardt's rather
11:40:32scorcheLlorean: no, but i havent touched them yet and will do that either tomorrow or monday...feel free to edit that page
11:40:33austriancoderendpoint0 has rx and tx enabled..hmmm
11:40:59MonkeyTamerseems to have an audible difference for me; gotthardt's commit that is
11:41:20markunMonkeyTamer: is it better?
11:41:32MonkeyTamerit's slightly louder
11:41:40MonkeyTamerI'm trying to notice the differences
11:41:48MonkeyTamersounds a bit cleaner
11:41:59MonkeyTameras in less static sounds
11:42:18LloreanMonkeyTamer: What's your volume at?
11:42:33LloreanWhat'd you have your bass at before?
11:42:45MonkeyTamerand the setting is 10 now
11:42:49MonkeyTamerwhatever that means
11:42:54LloreanIt means "10"
11:43:04LloreanThe codec's datasheets apparently don't say what it means either.
11:43:15MonkeyTamerI understand that, but in relation to Db I don't know how to relate
11:43:18MonkeyTameroh gotcha
11:43:24LloreanThey can't really be related, iiuc.
11:43:27amiconnYeah, 10 dunnothings. That's what I wouldn't like if I had a gigabeat
11:43:36LloreanYou get more bass boost on quieter portions, and less on louder.
11:43:38LloreanIt's dynamic.
11:44:03Lloreanpetur: I think it's just an arbitrary "how much? Oh, 10, okay"
11:44:10LloreanLike volume knobs that go from 0 to 20
11:44:59peturamiconn: where is our 'if the hardware supports it, we support it' idea gone to? If it makes the output sound cleaner and/or if users like it, why not offer it?
11:45:04MonkeyTamerthe difference between 0-15 is actually pretty small
11:45:23Lloreanpetur: The problem is, it prevents a user from manually setting bass as it is right now
11:45:40peturok, that has to change then
11:45:50Lloreanpetur: It should be a separate option (like some of the options the Recorder has for sound modification) rather than replacing "Bass"
11:45:58peturI agree
11:46:10peturdidn't know it replaced it
11:46:23 Join unibrother [0] (i=810d4899@gateway/web/cgi-irc/
11:46:31LloreanJust a second option, "Dynamic Bass" or something, and a mention in the manual that it overrides the main bass (assuming it does)
11:46:39LloreanOr whatever it's called
11:47:00scorcheadaptive =)
11:47:06LloreanAdaptive, there you go.
11:47:48LloreanBut for the moment, it might be nice if it were disabled before it screw up people's settings.
11:48:04LloreanIE, anyone who updates a build and goes "Where has my +16dB of bass gone, it sounds like a bag of shit now!"
11:48:07scorcheand it shouldnt be default anyway
11:48:22MonkeyTamerLlorean: haha
11:48:28MonkeyTamerit's not that noticeable
11:48:46LloreanMonkeyTamer: If you were one of those crazy people with bass on max, I bet it would be.
11:49:15MonkeyTamerfair enough, the setting at least doesn't sound like -6.0db though
11:49:21markunLlorean: according to wolfson it "usually sounds more pleasant to the human ear"
11:50:07markunso it probably is a violation of the rockbox BoS sound :)
11:51:01Lloreanmarkun: I bet it does sound more pleasant.
11:51:10*scorche hates ambiguous units with a passion
11:51:12LloreanIt just shouldn't replace the main Bass option.
11:51:18amiconnpetur: The point is to offer it, not to make it the standard
11:51:29LloreanWe start with Bass at 0dB boost anywya
11:51:37Zagorif noone else does, I'll take a whack at it after lunch
11:51:44LloreanSo with it, we start with Bass and aBass at 0dB and "0" (or Off) respectively
11:51:48peturamiconn: I think we agree then ;)
11:51:49MonkeyTamerwell to me it sounds like the Bass isn't so overwhelming, as in not competing with the vocals so much
11:51:59LloreanDoes aBass supercede the normal bass setting?
11:52:12markunyes, you can't use both
11:52:19amiconnpetur: The commit made it the standard, with no chance to use dB
11:52:24Nick_Brackley_1is this 'adaptive bass' available on other players?
11:53:24Lloreanmarkun: Then I wonder if it's possible to make it clear just in the menus that one disables the other
11:54:11scorcheLlorean: well, since everyone reads the manuals as they should, i dont think that would be the issue
11:54:26MonkeyTamerof course
11:54:27markunNick_Brackley_1: probably, but I don't have the other wolfson datasheets at hand here
11:54:37 Join PaulJam [0] (
11:54:48austriancoderamiconn: do have some time to help me finding out why tx in my usb stuff does not work?
11:54:48LloreanCan settings be mutually exclusive? For example, setting aBass to "3" automatically resets Bass to 0, and vice versa?
11:54:51Nick_Brackley_1Llorean: add a new menu item "bass mode −−-> adaptive or linear
11:55:20LloreanNick_Brackley_1: That'd work too.
11:55:51Nick_Brackley_1Llorean: i think that would be the easiest to understand from a users perspective
11:55:52Lloreanamiconn: You might know this. Any reason "make clean" doesn't remove rockbox-info.txt, rockbox.ucl and sysfont.h for recorder builds?
11:56:11PaulJamWouldn't it be best to make it like the peakmeter setting. so you have one setting "bass type" with fixed and adaptive , and one setting with the actual value.
11:56:21LloreanPaulJam: That's what Nick said
11:56:34markunbut the setting uses different units depending on the type
11:56:45LloreanYep, that's where it'd get confusing for a user.
11:57:07LloreanEven if they aren't labelled "dB" once mode is changed to Adaptive, I'm sure some will think it is having seen the old values
11:58:01scorcheare the menus dynamic enough to change depending ont he value? (i havent looked into jdgordon's code)
11:59:32amiconnmarkun: The different peakmeter modes also use different units
11:59:39amiconnOne uses dB and one uses %
11:59:54markunLlorean, amiconn: do you think the other options of the DAC should be available too? bass filter cutoff and trebble filter cutoff
11:59:55MonkeyTamerit sounds like a better implementation of the bass from a listening standpoint; the lowest setting is not ridiculously low, nor does the max setting have excessive bass (as in blocking in the vocals kind of excessivenss)
12:00:12Lloreanmarkun: I think I already agreed those should be. If not, I do.
12:01:01amiconnI'll make those settings available on G5 as well
12:01:20amiconn(or rather, keep them available when killing the hw eq screen)
12:01:21markunas long as we have good defaults it doesn't hurt to add them
12:01:41Nick_Brackley_1i think the adaptive setting needs to be assigned some kind of a unit, to fully remove the confusion
12:01:57LloreanMonkeyTamer: Better is probably subjective. If you're looking to make up a deficiency in hardware that reduces the Bass, you probably want a flat amount applied (and the Bass setting is "cheaper" than the EQ)
12:02:05markunsome people were complaining about lack of bass in rockbox compared to the ipod OF, maybe that's because of these settings
12:02:09LloreanNick_Brackley_1: You can't "assign" a unit when there is none
12:02:35LloreanThe best we could do is "Off, A little, Some, More, A respectable amount, Halfway, More than half, Even more than that, A good deal, Most" or some such
12:02:38LloreanWhich is kinda silly
12:02:45Nick_Brackley_1well then the unit needs to be discovered ;)
12:02:47markunor a percentage
12:02:50 Quit Thundercloud (Remote closed the connection)
12:02:50markun0 to 100
12:03:02markun(steps of 6.666.. :)
12:03:25LloreanOh, wait, 0
12:03:34Nick_Brackley_1i was thinking percentage before actually
12:03:51MonkeyTamerwhat I mean is that the transition between gradations, 0-15, feels like the increase flows, while the previous implementation sounds like steps; this is definitely my opinion and is subjective, I was just providing some input
12:03:56markunLlorean: no, you are right
12:04:24LloreanMonkeyTamer: That's probably because you're moving across a much smaller range, with a similar number of steps.
12:04:31MonkeyTamerI agree
12:04:54MonkeyTamerthe difference between 0-15 is definitely much lower than previously
12:05:25markunMonkeyTamer: but the difference should also depend on the volume, at a lower volume the steps should be higher
12:05:41markunbigger I mean
12:05:52MonkeyTamerI'll play with it
12:05:57peturmarkun: no, you were right
12:06:30markunpetur: but there is also 0 (off)
12:06:55peturyes, but there are only 15 steps between 0 and 15
12:07:04markundamn :)
12:07:29LloreanSo, approximate percentages (7%, 13%) etc could work.
12:07:38markunyes, I would say so
12:07:45LloreanBut it's unclear then why we don't offer the in-between values.
12:07:56markunof course we will get those questions :)
12:08:15peturLlorean: we had the same discussion when implementing the RGB settings
12:08:27linuxstbDoes the datasheet describe what 0..15 mean?
12:08:45Lloreanpetur: True
12:08:54markunlinuxstb: no, just 1 (min) and 15 (max)
12:09:13Nick_Brackley_1Llorean: all values could be chosen and just rounded to the nearest coresponding one?
12:09:35LloreanNick_Brackley_1: That's a lot of values that don't change the sound, though.
12:09:46LloreanEven though apparently we do that with volume on certain targets.
12:10:15LloreanI think % is the best.
12:10:25LloreanOr just unitless
12:10:34LloreanWith a note in the manual saying that it doesn't correspond to any units.
12:10:44*Llorean can't really decide which he likes.
12:10:44Nick_Brackley_1Llorean: a sliding scale like the graphical eq?
12:11:07LloreanNick_Brackley_1: I personally would like to have as few of those as possible in the menus.
12:11:13linuxstbmarkun: Yes, I'm just reading it - so 0 is "off"
12:11:14scorcheLlorean: i think to a user, having a percentage that has large gaps that arent selectable is more confusing than units
12:11:42markunwe should ask a user!
12:11:42Lloreanscorche: I don't think it is, but I think it will become more confusing if say... the feature is implemented on another chip.
12:11:57Lloreanscorche: Say a chip that has the scale be from 0 to 7 or something
12:12:21linuxstbI would say just call the "off" and 1-15
12:12:25linuxstb^call them
12:12:30LloreanI think I agree
12:12:35*scorche still hates ambiguous units with a passion
12:13:05linuxstbI agree, but there's nothing we can do - there is no unit, or at least the datasheet doesn't describe it.
12:13:07markunscorche: how passionate are you about them? Would you stop working on rockbox?
12:13:17Nick_Brackley_1i think that will keep it as simple as possible
12:13:23scorchemarkun: dont make me picket
12:13:50LloreanDo all of our numbered lists that have units, display the units?
12:14:09MonkeyTamersheesh Wolfson, if you're going to provide public documentation, at least use non-ambiguous units
12:14:11scorchei think we should display them as ambiguous units, but i am referring to hating the manufacturers rather than our display =p
12:14:16markunfrom picket: a detached body of soldiers serving to guard an army from surprise
12:15:00MonkeyTamertreble and width do
12:15:18MonkeyTamereverything else in sound has a unit
12:15:24LloreanWhat I mean is, as long as we're consistent (whenever a unit isn't shown, it means there is no unit, and we don't have implied units somewhere) then I'm quite happy with no units or anything
12:15:30markunMonkeyTamer: which units could they have chosen?
12:15:32scorchemarkun: slang for a protest typically done outside of a building named after the wood used to mount the signs that people are displaying on
12:15:36LloreanMonkeyTamer: Treble doesn't have a unit?
12:15:44MonkeyTamerno, it does
12:15:46 Nick Tr1ckY^Gone is now known as Tr1ckY (
12:15:56MonkeyTamerI was listing the ones that do, which are pretty much all except bass no
12:16:21MonkeyTamerand of course on the EQ the 'Q'
12:16:43LloreanBut Q has a set definition for what the value means.
12:16:43 Quit unibrother ("CGI:IRC (EOF)")
12:16:47scorchean example of usage:
12:17:07MonkeyTamerso then bass is the only undefinable one
12:17:15LloreanLike if you had a menu setting "Kilometers Per Hour" units wouldn't need a display under it. ;)
12:17:43MonkeyTamerI really don't have a good concept of what Q really is to be honest
12:17:52LloreanI think as long as the only time no units are listed is when there really are no units, I think it's fine to do it.
12:19:06MonkeyTamerif one tried to put a value to it, the difference between 1 and 15 audibly compared to the old one during a bass line subjectively sounds around 3.0Db
12:20:12markunMonkeyTamer: I believe it's dB btw
12:20:30 Join Wiwie [0] (
12:20:33MonkeyTamerI was hoping that would be disregarded :p
12:20:55MonkeyTamerI noticed it after typing but didn't feel like correcting myself
12:21:00markunwell, you made the same mistake earlier so now I thought it couldn't be a typo
12:21:22MonkeyTamerprrobably exhaustion for me
12:21:28MonkeyTamerits 3:21 AM here
12:21:32LloreanThe pinky's just one character ahead of where it needs to be.
12:21:36MonkeyTamerand I've been up since 6
12:21:39scorchelinuxstb: re:, we still think that the element of us manually going through themes would be best if solely for the copyrighted material issue
12:21:48Lloreanscorche: He agreed
12:22:09*MonkeyTamer is confused by scoreche's link
12:22:10Lloreanscorche: If you read what he said, it's basically "One of us looks at the screenshot, then clicks Yes/No" if it were fully scripted.
12:22:15markunMonkeyTamer: just get some sleep man :) We'll be still here when you wake up
12:22:26MonkeyTamerof course
12:22:54MonkeyTameroh well, I'll be sure to find out the conclusion of the bass debate eventually
12:23:00MonkeyTamerg'nite all
12:23:08 Quit MonkeyTamer ("ChatZilla [Firefox]")
12:23:18scorchelinuxstb: yes..that was actually how i first envisioned the site, but when i got the site dumped into my lap, i figured that it would be easiest (and that the site would be up quickest) if i just did things this way
12:24:32scorchei would love to move to that format, but as i said, others would have to do it as i am incapable of making it that advanced, and the sooner we get the site up, the better...perhaps we can just get it up this way until we can move to that route
12:25:02nekura42I'm out as well.. night all
12:25:27 Part nekura42
12:25:42scorcheand the "small number of admins" atm (and in the future according to current plans) is just redbreva and i, although i dont have any issues with making more (and doubt he would as well
12:25:57*scorche hopes linuxstb is listening and /me isnt typing this out in vain
12:26:16 Join t0mas [0] (n=tomas@rockbox/developer/t0mas)
12:26:42markunhi t0mas!
12:26:48markunlong time
12:26:53linuxstbscorche: Why the rush?
12:27:19t0masreally long time
12:27:38t0masbut now end of university year.. and 2 weeks no work apointments..
12:27:49t0masso time to do some fun stuff again :D
12:28:15linuxstbI've also missed the background to this - is the plan for it to still be hosted where it is, but RaeNye has agreed to give more people access to the server to help admin it?
12:29:23scorchelinuxstb: it has needed to be done a while ago, and the longer we wait, the more themes have to be re-submitted (as a fresh start would be better imo than pruning the current repository) well, this interfaces with rbutil, and i would like to be able to have it point to officially approved themes rather than the current including copyrited material themes as it is now
12:29:34markunnice firmware update
12:29:35Lloreanlinuxstb: The plan is to make the official host for WPSes, and drop the wiki portion. Since it allows both more organization, and integration with RBUtil nicely
12:29:43markunstarts to look a lot more like rockbox :)
12:29:44scorcheyes...when this goes live, wll point to is
12:29:58scorcheand the owner is RedBreva...not RaeNye
12:30:15*linuxstb has a bad memory for nicks
12:30:57linuxstbOK, I guess if you're happy to do the manual admin, it makes sense to make it live asap and then improve the background stuff.
12:31:00Nick_Brackley_1so who is going to write this new site?
12:31:23linuxstbWouldn't "" be a better name though?
12:31:25 Quit GodEater (Remote closed the connection)
12:31:34linuxstbJust so people know it's official.
12:31:49*markun agrees
12:31:53*Llorean nods too.
12:31:58scorchei definitely wouldnt mind us eventually moving towards the route you described (especially as that is how i envisioned, but could not implement)
12:32:10scorchelinuxstb: the plan was to have point to
12:32:18 Join Nick_Brackley [0] (i=3aa67da1@gateway/web/cgi-irc/
12:32:18 Quit Nick_Brackley_1 ("CGI:IRC (EOF)")
12:32:21linuxstbSeems an odd plan...
12:32:36scorcheas the swedes would not like FTP access to their servers likely
12:32:54linuxstbI'm just talking about the name of the site being changed, not the server.
12:33:44 Quit webguest76 ("CGI:IRC (EOF)")
12:33:58scorchei suppose we could, although that domain is going to waste =) (or redbreva can just keep that as his personal repository for grey-line themes we dont like)
12:34:11LloreanWell, you can always just have both domains work
12:34:20GodEater_that would be better
12:34:45linuxstbIt seems a shame to lose all the themes on the current site - maybe the style of the existing could be changed (so it doesn't look official), and then a new, empty site started on
12:35:04scorchelinuxstb: that was part of the plan
12:35:29scorchehe was going to move the current r-t to a subdomain and maintian it completely separate from ours
12:35:47scorchegod, i cant type...
12:36:17linuxstbBTW, is it a Linux server?
12:36:30scorcheit should be
12:36:44scorcheyes, it is
12:37:06Nick_Brackleydo you have someone to write the new site?
12:37:40scorchewell, this past week, i have been getting the site ready for deployment (and it currently is except for the guidelines page)
12:37:55linuxstbWhat do you think about putting the PHP in SVN?
12:38:06*Llorean thinks it should be.
12:38:07scorchei dont see a problem with having it in svn though and having linuxstb or someone fluent in PHP edit it
12:38:12scorchei was typing that out =P
12:39:04markuntoo bad we don't have dongs anymore for suggesting we use ASP.NET
12:39:04linuxstbBut you only have ftp access to the server, not ssh?
12:39:38scorchewe have both, but i have not needed ssh access and so didnt ask for it
12:39:51scorchewhat will ssh access be needed for?
12:39:55linuxstbI'm just thinking we may want to run housekeeping scripts on the server
12:40:05linuxstbe.g. if a WPS tag changes, run a script to fix all WPSs
12:40:22LloreanThat would be handy.
12:40:25 Join Thundercloud [0] (
12:40:26scorchei would have to run that by redbreva, but i dont see why not
12:40:30safetydanmarkun, I can suggest Java and JSF it that helps :)
12:40:38markuna lot!
12:40:41LloreanOr if a font is added to SVN, check exiting WPSes for it and remove it from the archive, etc?
12:40:48linuxstbBut I guess we could have "export all WPSs" and "import all WPSs" admin options to allow that to be done offline.
12:40:51Nick_Brackleyi'm fluent in php and sql
12:41:07scorchelinuxstb: i would much rather have that done offline if possible, yes
12:41:21*linuxstb is just used to poking live servers
12:41:26Lloreanscorche: Online would be a good deal faster as long as it's a low-risk script
12:41:57 Join freqmod [0] (
12:42:10scorchetrue...i just dont know how picky 1and1 is...they got mad at me for a cgi script on once...
12:42:33linuxstbAre the WPSs currently stored as zip files on the server?
12:42:34 Quit t0mas ("brb, changing hotspots")
12:43:15scorchethey are currently stored as zip files, a txt file to pull the info from used to generate the page, and 1 (or 2) pngs to display when the page is built
12:43:28scorcheit is (of course) built on each request
12:44:02scorchethe txt file has items like the submitters name, description, etc
12:45:59linuxstbAre any automated checks currently done on the uploaded zip file?
12:46:07scorchenot currently
12:46:56safetydanamiconn, have you already done the work of removing the hardware equalizer menu?
12:49:20linuxstbscorche, Llorean: What about themes that need patches? Are they going to be hosted?
12:49:52scorchelinuxstb: that was the plan...i have a subdirectory with the separate repository, rules, and fields all set up
12:49:53 Quit Nick_Brackley ("CGI:IRC (EOF)")
12:49:57 Join Nick_Brackley [0] (i=3aa67da1@gateway/web/cgi-irc/
12:50:15LloreanHosted, but completely separated.
12:50:19scorcheactually i take that isnt *quite* finished yet
12:50:42LloreanAnd when automatic 'validation' is working, they won't be subjected to the WPS parser test, of course.
12:50:52scorchelinuxstb: perhaps just looking at it will answer some questions is the version i have been working with
12:51:56linuxstbLlorean: I think if we host them we should do it properly - i.e. have a list of specific patches that are required by the theme, and build versions of check_wps with those patches. Although I'm not sure how many patches we're talking about.
12:52:27scorchemost links point to the www version as i changed that to be civil with the 2 (basically the same file with minute changes) files for the patches and non-patches site
12:52:48GodEater_does anyone know if the mAh value for the battery is quoted for an iPod 80GB 5.5Gs anywhere ?
12:52:55Lloreanlinuxstb: The plan was to require a list of flyspray numbers given with the Theme, but not to do the parse-check, since either we just need a parser with every possible patch installed, or a million parsers (plus we have to keep those up to date as well)
12:52:55 Join lunte [0] (
12:53:11scorcheit isnt worth parsing those imo
12:53:14linuxstbscorche: Maybe "For Official Rockbox" -> "For any Rockbox" (if that's true)
12:53:37scorchelinuxstb: ah...good idea..i was at a loss for those titles
12:53:49lunterun rockbox on a odys MP X50?
12:54:02Lloreanlunte: Rockbox only runs on those player listed on the front page
12:54:24midkayscorche: you were not at a loss! what about the names i gave you. :(
12:54:47scorchemidkay: i used those...pretty much
12:54:52markunlunte: it's not impossible to support it
12:55:00markunbut takes a lot of dedication
12:55:27midkayscorche: right, and suddenly you're at a loss for such names? :(
12:55:35Lloreanlinuxstb: I wouldn't mind a parse-test of some sort for the patched ones, but I'm not sure how feasible setting a reliable one up is (and maintaining it as patches change their tag requirements, etc)
12:55:49scorchemidkay: i was describing my state before...
12:56:00LloreanIt seems easier to make it clear that those have no guarantee, "unsupported" in the same way unsupported builds are.
12:56:21midkayscorche: but you said "good idea" as if you were accepting them.
12:56:22linuxstbLlorean: I think it depends on how many patches there are, and how frequently they tend to change. I haven't been following that area at all.
12:56:50Lloreanlinuxstb: Well, in terms of patches that are in our tracker, there may only be about 4.
12:56:59linuxstbBut e.g. if there are a small (3 or 4) number of patches that cover the vast majority of WPSs, then we could only host those.
12:57:23amiconnsafetydan: nope
12:57:26scorcheit isnt worth the effort at all, especially as patches change, some get accepted, features get added, etc
12:57:35LloreanAlbum Art, which has had set syntax for a while, Custom Scroll Margins, one for a modified progress bar (if I recall), and I think there was one other I can't remember.
12:57:55GodEater_Llorean: isn't there a multiple-font one ?
12:58:02Nick_Brackleymultifont, customline
12:58:25LloreanWe could do a parser for AA+Scroll+Multifont+Customline, then just flag it as pass/fail for that test, but let them through either way?
12:58:33GodEater_what does the progress bar one do ?
12:58:50scorcheregardless, they are constantly in flux and i really dont think parsing them will do any good (as they are better left like the unsupported builds)...lets just focus on the main site for now? =P
12:58:53LloreanGodEater_: Lets you move a single image across the screen, rather than filling up a bar, iiuc.
12:58:55Nick_Brackleyseems like alot of maintance for not much gain imo
12:59:26scorcheexactly...i really dont think it is worth it on the themes that require patches
12:59:32safetydanamiconn, if you've got better things to do I can always try and hook that up. I'm stuck on my other patches anyway.
12:59:37linuxstbMaking a list of these patches would also be useful in terms of deciding why they're not in SVN, and seeing if they can be.
13:00:27scorchein the future, one would be able to just browse the page and see what numbers people are using
13:00:41scorche(near future, as in when the themes are submitted)
13:02:16Lloreanlinuxstb: Multifont requires a good chunk of memory, Album Art = MOB, Scroll Margins = Viewports, and I believe CustomLine = Viewports as well
13:03:43LloreanIn regards to multifont, I've wondered aloud at one point if it'd be possible to simply pre-render the strings for the WPS and any given song, and (for the song) store it in the buffer, so that a full font isn't necessary.
13:03:44***Saving seen data "./dancer.seen"
13:05:09LloreanOf course, it wouldn't work well with say, changing WPSes mid-playlist.
13:05:47amiconnLlorean: The unicode layer does lru caching anyway. Fonts are rarely loaded completely (some even can't be)
13:06:01*amiconn wonders what happened to markun's rework ideas
13:06:55Lloreanamiconn: So in the case of multifont support for WPSes, can the caching simply be extended to glyphs from separate fonts somehow?
13:07:59 Join t0mas [0] (n=Tomas@rockbox/developer/t0mas)
13:08:00t0masback :)
13:08:13scorchet0mas: for good? =P
13:08:23t0mashehe.. who knows? :P
13:10:00markunamiconn: I know.. lazyness..
13:12:58 Join Entasis [0] (
13:13:05linuxstbLlorean: I'm just thinking that it would be good to try and focus development effort on the popular WPS features - to get more people using unofficial builds. A lot of people seem to only want Rockbox for its theming capabilities...
13:13:27linuxstb^I meant more people using official builds...
13:13:42LloreanI assumed so. :)
13:13:44LloreanI agree
13:13:57 Quit miepchen^schlaf (Read error: 113 (No route to host))
13:14:05LloreanBut most of the popular theme features have been addressed one way or another, I think, and have a requirement for entry
13:14:10linuxstbe.g. maybe just a wiki page describing the patches, and what the problems are, and maybe ideas for solutions.
13:14:16LloreanThat would be nice.
13:14:52linuxstbWhat's the "custom line" patch? get back to the site (i would like to get to bed soon), should we proceed with the plan on making the current version (with my modifications) live and when the backend is revamped to eliminate the human equation (linuxstb or someone fluent in PHP), we make that live
13:15:19Lloreanlinuxstb: I think it's Y positioning in WPses for lines?
13:15:42linuxstbOK, then yes, viewports will meet that need.
13:15:52LloreanI'm not 100% sure that's that patch though.
13:16:37linuxstbI also remember the custom line height patch, which could still be useful with viewports (as IIUC, viewports will still allow line-based text within them)
13:16:40Lloreanscorche: I do agree it'd probably be nice to try to get it up, with maybe a warning posted that theme approvals might be "slow" until the system is improved.
13:17:06Lloreanlinuxstb: Yes, I was quite fond of that patch, but apparently it didn't like scrolling (or something, I remember there being something that turned out not to work)
13:17:07scorchelinuxstb: your thoughts?...getting the site into svn is included in that plan
13:17:51linuxstbscorche: I think the most important thing is the (C) issue - anything else can be changed in the future.
13:17:57 Join otih_ [0] (
13:18:18linuxstbe.g. we could accept unofficial WPSs now, and then run them through check_wps in the future to confirm what unofficial tags they're using.
13:18:23*scorche takes that as a "sure"
13:19:10linuxstbscorche: Yes, I've no objections go going live asap
13:19:50scorcheand would you mind editing the pages to do as we wish, or shall i hunt for someone else (no strict timespan on that)
13:20:34linuxstbLlorean: Maybe something like "FutureWPSFeatures" - a list of features generally agreed to be desirable, along with links to patches that implement them, and comments about why the patch implements it in the wrong way (and hence is not in SVN).
13:21:47Lloreanlinuxstb: Sound good. Should I request in some of the more popular unsupported build pages that they feel free to populate the page with "Theme related patches" that can then have relevant When/Why info filled in by someone in the know?
13:23:48LloreanI'll put Album Art up as a sample even.
13:23:59linuxstbscorche: There's room for more than one person - so if others are interested, let them go ahead. I'm not particularly fond of PHP hacking, but will be happy to do some work on it - I'm thinking about the validation of the uploads and how check_wps could be used.
13:24:31scorchelinuxstb: alright...perhaps i shall make a wiki page on it later
13:25:33Nick_Brackleyim interested in php hacking and have experience, if you decide you need a hand
13:25:34scorchei wouldnt really call the code of the page "pretty", but it works =)
13:26:15scorcheNick_Brackley: i will keep that in mind if you plan on staying around here for a bit...thanks =)
13:26:16linuxstbLlorean: That could work, but we need to be careful it doesn't turn into a free-for-all page where people submit feature requests. I was thinking of something more like a sort of developer-related "to-do" and technical discussion page.
13:26:44scorchei would think a wiki page would do that justice, and am not sure why we need to go into the forums...
13:27:00linuxstbscorche: Is the PHP very hard-coded for the existing site, or should it be easy for people to install locally?
13:27:19Nick_Brackleyscorche: yeah i'm a long time user of rockbox and really want to contribute something back, im here for the long haul :)
13:27:19Lloreanlinuxstb: ?
13:28:11scorchelinuxstb: it used to be more local, but i have been making it a bit more hard coded at least for the moment while i was dealing with a few versions of basically identical pages
13:28:19scorchei can make it more local, i suppose
13:28:38Lloreanlinuxstb: Maybe only those with an existing patch, no feature requests?
13:29:03linuxstbI normally do something like have an "ini.php" file defining site-local variables. So to use the site locally, you swap the ini.php file.
13:30:13scorchelinuxstb: i really have tried to stay away from php in the past and now it is starting to bite me...
13:30:30linuxstbLlorean: Not necessarily - e.g. viewports could be described as the feature, and things like scrolling-margins and customline could be referenced as partial patches for that feature.
13:31:12Lloreanlinuxstb: "Only non-rejected patches, or features confirmed by developers" then?
13:31:28LloreanI'm just trying to put together guidelines for what should be on the page.
13:31:45 Quit BigMac (Read error: 110 (Connection timed out))
13:32:06 Join Arathis [0] (
13:32:15 Quit otih (Read error: 110 (Connection timed out))
13:32:31 Join maffe [0] (
13:33:04linuxstbI was thinking of an introductory sentence saying something like "This page describes WPS features that are considered by the developers to be desirable for future versions of Rockbox, and the current state of their implementation".
13:33:47Nick_Brackleylinuxstb:nailed it
13:34:41safetydanSomeone want to check I haven't broken anything with this:
13:35:26safetydanPatch basically removes the hardware eq and changes the audiohw_set_bass in wm8758.c to use the EQ1 for bass and EQ5 for treble
13:35:36safetydancutoff frequencies are currently hard coded
13:36:33Lloreanlinuxstb: Go look at the page now?
13:37:37 Join spiorf_ [0] (
13:40:10Nick_BrackleyLlorean: just an idea, would indenting the Scroll Margins info distinguish it better as a sub-catagory of sorts?
13:42:32LloreanNick_Brackley: Doesn't seem to be a good solution for indent other than manually typing a lot of spaces. How does HR look?
13:43:35 Quit Wiwie ("Miranda IM! Smaller, Faster, Easier.")
13:44:25Nick_BrackleyLlorean: I like the HR it separates the sections nicely, although i still think something needs to be done for the sub items, smaller font, italic or something along those lines?
13:44:32 Quit spiorf (Read error: 110 (Connection timed out))
13:44:41LloreanNick_Brackley: Technically it is a smaller font in the header.
13:44:55LloreanI'm just using the default twiki syntax for "sub-item"
13:45:20Nick_BrackleyLlorean: ok no worries
13:45:30scorcheNick_Brackley: you can also see quite clear the difference in the TOC
13:46:38Nick_Brackleyscorche: yes and that was my reasoning for a change in sytle/formating for sub-items, to keep with the TOC
13:46:56scorchethe TOC is automatically generated depending on the page
13:47:27LloreanAdded one little line to the intro area
13:47:49LloreanThink it's okay if I request in a few build threads that users feel free to add to that page, within the guidelines?
13:48:21Nick_Brackleyah ok. makes sense
13:55:38 Quit safetydan ("Ex-Chat")
14:01:25 Quit gotthardt ("adios!")
14:04:26linuxstbLlorean: Your page isn't what I have in mind now - I've moved from "here are the patches, this is why they're not included" to "here are the features we want, and here is the current status of any effort to implement them".
14:05:31Lloreanlinuxstb: Does mine come off as "here's why they're not wanted"?
14:06:00linuxstbIMO it's too focused on existing patches, rather than taking a step back and thinking "what features do we want?"
14:06:32LloreanHow do we do that without it just becoming a secondary feature request site?
14:06:52linuxstbWith my suggested introductory paragraph...
14:07:09linuxstbi.e. make it clear that it's features developers want - so users don't think they should be adding features there.
14:07:29 Join Wiwie [0] (
14:07:49Lloreanlinuxstb: Where is your suggested introductory paragraph?
14:08:09linuxstb25 minutes ago in the logs (just one sentence, not a para)
14:08:17LloreanOh, just that sentence
14:08:32linuxstbI think that's all the intro needs to say.
14:08:51Zagorregarding the adaptive bass on gigabeat: what do you think about simply having two settings. and when one is set the other is nullified? slightly confusing, but it would avoid the need for a "switch" where you select that bass mode
14:09:50linuxstbI think that would be slightly confusing...
14:10:03linuxstbSo changing one setting would change the value of the other?
14:10:10Zagoryes, to 0
14:10:22ZagorI know, it's not ideal. but neither is a "switch"
14:10:24linuxstbI'm not keen on settings being changed behind the scenes.
14:10:25Lloreanlinuxstb: Your sentence, + "Please only post WPS and themeing features that have an existing patch that is working toward inclusion, or a feature that has been confirmed as planned by developers." ?
14:11:08LloreanHow 'bout, if you try to change one, and the other is a Non-Zero, a splash saying "Bass must be deactivated"
14:11:10linuxstbLlorean: I think that's implied by the first sentence, and it shouldn't depend on the existence of a patch.
14:11:14LloreanOr "Bass must be 0"
14:11:20LloreanRather than an automated change.
14:11:29Lloreanlinuxstb: That's why the "or" is in there.
14:12:10LloreanBut I guess you're right
14:12:20linuxstbI always think less is more...
14:12:53ZagorLlorean: I think that would actually cause even more confusion :-) "why won't rockbox let me set any bass?"
14:13:00LloreanAlright, the intro is three sentences. Your sentence. One mentioning that patches can be grouped as parts of a feature, and the one saying "This isn't the place for debate"
14:13:38LloreanZagor: Well, I think I actually like the "switch" best, personally
14:15:14 Join donutman25 [0] (n=chatzill@
14:15:41Nick_BrackleyZagor: I think switch is the simplest easy to understand solution, too
14:15:49linuxstbWhat about having the "Bass" setting change (from dB to 1-15) depending on what you've selected as the current bass mode? So they would be two settings, but only one ever visible in the UI at any one time.
14:16:36Zagorlinuxstb: good idea.
14:16:44Nick_Brackleylinuxstb: yeah sounds solid
14:16:50LloreanI thought that's what was meant by having a switch
14:17:30linuxstbYes, you would have a switch, but hide the non-active bass setting. (and call them both "bass")
14:17:31ZagorLlorean: actually my first idea of the switch was to select which setting is used. but linuxstb's variant is much better.
14:18:19Nick_Brackleyi think that is the ideal way to go linuxstb
14:19:04linuxstbHopefully the settings menus can cope with dynamic items...
14:19:25Nick_Brackleylinuxstb: yeah are there any in there currently?
14:19:35LloreanI thought someone said the peakmeter setting does.
14:19:52LloreanThe range changes if you're using linear vs logarithmic or whatever the options were.
14:20:09LloreanWhich is kinda why I thought we were talking about having the values offered change already. ;)
14:20:10linuxstbSounds the same as the bass issue then.
14:20:26 Quit lunte ("werde zurck in den ALK-Olymp gehen")
14:20:46Nick_Brackleyyeah so as long as we keep it consistent it should be ok
14:26:25 Nick ubuntu__ is now known as chrisjs169_ (
14:27:37Zagorhmm, our sound settings are a bit off from wolfson's recommendations
14:28:52Zagorspecifically the cut-off frequencies
14:30:42 Join Lynx_ [0] (
14:30:57Zagorhas anyone here looked any at it? we have the bass cut-off hard-coded at 200Hz instead of the recommended 130Hz for instance.
14:31:30Zagorwith the "if the hardware supports it, so should we" philosophy this should probably be user-settable as well...
14:32:53LloreanZagor: I believe Amiconn mentioned we can have it user-settable on the UDA1380s as well.
14:33:07pixelmaI believe markun mentioned something about it before not sure if that's what meant...
14:33:57 Part pixelma
14:35:31markunLlorean: so you we planning on adding the options, right?
14:35:52FebsIt is user-configuration on the UDA1380. I don't remember the available cut-off frequencies offhand, but I do remember seeing it in the datasheet.
14:36:14Febsuser-configuration/potentially user-configurable
14:36:47Lloreanmarkun: I wasn't planning on adding anything but an opinion. Honestly, the only time I use sound altering controls is to try to fix some bad audiobooks ripped from old tapes.
14:37:06LloreanI don't remember who mentioned interest in doing it.
14:40:33FebsUDA1380 bass boost edge frequency can be either 250 Hz or 300 Hz, treble boost edge frequency can be either 1.5 kHz or 3.0 kHz.
14:42:30markunZagor: these values are at 48kHz, can we just scale them for 44.1? (119 and 184)
14:42:42FebsAt 250 Hz, the gain range is 0 to 18 dB. At 300 Hz, the gain range is 0 to 24 dB.
14:42:52Zagormarkun: yes
14:43:13Zagor"All cut-off frequencies change proportionally with the DAC sample rate."
14:44:02amiconnThe UDA datasheet is somewhat unclear on this
14:44:37 Join norbusan [0] (
14:44:44 Part norbusan
14:44:58 Quit chrisjs169_ (Read error: 104 (Connection reset by peer))
14:45:05amiconnThey mention that the cutoffs change with sample rate, but are fixed otherwise. A bit later they mention this off/minimum/maximum tone control setting, saying "fc" (whatever that is) change with it
14:45:40amiconnThose "fc" could be the cutoffs, but they have somewhat odd values for treble/bass cutoffs....
14:45:55amiconn(see log from this morning)
14:46:11ZagorI just noticed that the mdb_* settings are included for all targets...
14:49:38 Join Soul-Slayer [0] (n=Administ@
14:51:48 Join midgey_ [0] (
14:51:48 Quit midgey (Read error: 104 (Connection reset by peer))
14:52:42 Quit SirFunk (Success)
14:57:58 Join Xerion [0] (
15:00:06 Quit HellDragon (Connection timed out)
15:01:02Zagorthe menu system has changed a lot since I looked at it last :)
15:02:09ZagorI guess I should look at the code more than twice a year...
15:03:01linuxstbLlorean: This is the approach I had in mind for the FutureWPSFeatures page -
15:03:45***Saving seen data "./dancer.seen"
15:05:05 Quit midgey_ (Read error: 104 (Connection reset by peer))
15:05:29 Quit Wiwie ("Miranda IM! Smaller, Faster, Easier.")
15:06:24LloreanSo, basically just a "Status" page for UI Development?
15:06:31LloreanOr, specifically WPS
15:07:12 Join midgey [0] (
15:08:23 Join The-Compiler [0] (
15:08:56 Part Nick_Brackley
15:09:46 Join Wiwie [0] (
15:09:51 Join webguest55 [0] (i=c023110a@gateway/web/cgi-irc/
15:10:32 Join Nico_P [0] (
15:13:02 Quit Wiwie (Client Quit)
15:19:45 Join aliask [0] (
15:21:27 Join midgey_ [0] (
15:25:25 Join My_Sic [0] (
15:26:03 Quit The-Compiler (Connection timed out)
15:31:48 Quit PaulJam (".")
15:32:04[Ray]austriancoder: do you happen to have linux2rb.h floating around somehwere? seems to be missing from your patch
15:32:39austriancoderupps.. will add it in the next one
15:33:23 Join SirFunk [0] (
15:35:25 Quit midgey (Read error: 110 (Connection timed out))
15:39:16 Join midgey [0] (
15:40:33 Join excitatory [0] (
15:41:52excitatoryso i've been using rockbox on my 2gb nano for a year now and love it to death.. ..but obviously 2gb isn't exactly cutting it.. so i'm ready to buy a new device.. if money was no object, what rockbox-compatible device would you choose?
15:41:52linuxstbLlorean: Yes, I guess you could call it a status page - listing the features that we want, how we want them to be implemented, and what efforts (if any) are ongoing to implement them.
15:42:16Lloreanlinuxstb: Go ahead and replace it then. I just misunderstood.
15:42:33Lloreanexcitatory: Depends pretty heavily on what features you require.
15:42:41LloreanI'd choose an H140 with an 80gb drive.
15:42:47linuxstbexcitatory: There's no single best player - everything is a compromise. Do you want flash or disk, fm radio, recording, very long battery life, colour LCD, ....
15:43:10LloreanI think coming from a Nano, if you just want "Bigger, better, and more battery life" the Gigabeat F/X series are where to look.
15:43:22excitatoryLlorean: i need space, and the ability to play flac/ogg. i guess i'm interested in video, but have never played around with it outside of my pc
15:43:40excitatory..but video isn't necessary
15:43:51linuxstbFM radio or recording/
15:43:59linuxstb(the Gigabeat F lacks those)
15:44:11excitatoryi would take the radio if it were available, but it's not imperative.
15:44:34linuxstbIs 15h-20h battery life good enough, or do you want more?
15:45:02linuxstbexcitatory: This page may help -
15:45:53excitatorywell, i guess i wouldn't mind the most battery life available.. my 7-8 hrs on my nano is killing me.
15:47:07 Quit midgey ()
15:47:48excitatoryok, thank you all :D i haven't been here in a while..
15:49:59excitatoryare you guys getting swamped with iphone questions?
15:50:24linuxstbWhat's that?
15:50:38LloreanI think most of them happened before the release.
15:51:04 Join low_light [0] (i=c730180a@gateway/web/cgi-irc/
15:51:24*GodEater_ watched an iphone get put in a blender this morning
15:52:15excitatorythat one, and future multi-touch enabled devices seem like they're going to pose quite the challenge to hack/reverse-engineer..
15:52:33 Quit moos (Read error: 110 (Connection timed out))
15:52:47Zagorexcitatory: multitouch is not a problem. hard encryption is.
15:52:52Zagorand apple is not the first to do it
15:52:52excitatoryGodEater_: yea, i still need to view it
15:54:20 Join pepie34 [0] (
15:55:29excitatoryZagor: indeed. i guess the problem i was thinking of, is that rockbox would have to incorporate all sorts of new features in order to utilize a multi-touch (or even touch?) technology.
15:55:43 Quit midgey_ (Read error: 110 (Connection timed out))
15:55:46 Part austriancoder ("Kopete 0.12.4 :")
15:55:57excitatorysince to my knowledge, none of the compatible devices are any 'touch'
15:56:40Zagorexcitatory: touch is merely an x/y coordinate input from a serial port or similar. dual touch is two pairs of coordinates. it's nothing magic.
15:57:13LloreanNot to mention you don't have to implement touch at all if there's other input methods, it's not key to being able to use Rockbox.
15:57:14excitatorywell, it _is_ magic to someone who struggles to write a bash script :D
15:57:19Zagorit wouldn't be very difficult to for example keep track of which areas each menu entry occupies and allow direct-select
15:57:22 Join rogelio [0] (n=rogelio@
15:57:42 Quit rogelio (Remote closed the connection)
15:57:54excitatoryi suppose
15:58:04 Join chrisjs169_ [0] (
15:58:09Zagorbut as Llorean says, it's not certain touch is even wanted on a device you mostly keep in your pocket...
15:58:23chrisjs169_jhMikeS: you around?
15:58:25LloreanDirect select would be awkward with small fonts.
15:58:37 Quit XavierGr (Nick collision from services.)
15:58:40 Join XavierGr [0] (
15:58:55excitatoryvery true
15:59:11LloreanI'd just divide the screen into four triangles with bases on edges, and make them "Up" "Left" "Right" and "Down" like the joysticks, then make the center "Select"
15:59:27LloreanForget about anything really fancy.
15:59:32LloreanBut I'm a simple person.
15:59:49Zagorit could probably be fun in some plugins/games though
15:59:59LloreanOh, abolutely
16:00:31linuxstbOr just write a new list widget designed for touch screens, and use it instead of the existing implementation.
16:00:54LloreanI'm thinking one idea for next year's SoC (assuming no work gets done before then) is isolating the "Application" part of Rockbox to the point where it can be compiled with the intent of being a stand alone app (making it more suitable for port to interesting targets that can run applications)
16:01:02 Join midgey [0] (
16:01:16linuxstbLlorean: A SummerOfCode 2008 page... ;)
16:01:25LloreanBasically, a real UI for an "App" build, etc.
16:01:43linuxstbI agree that would be nice - I would like Rockbox as my desktop music player.
16:02:12Zagorlinuxstb: can't you do that with the simulator already?
16:02:19Zagornot very polished though :)
16:02:35peturI think the idea was to create a 'PC' (x86) target
16:03:08 Quit ansivirus (Remote closed the connection)
16:03:49linuxstbOr multiple ones - you could use Qt, GTK, ncurses, SDL, whatever for the UI.
16:04:11 Quit Zagor ("Client exiting")
16:06:04linuxstbLlorean: But yes, it's a perfect SoC suggestion - something everyone wants, but no-one will ever actually get round to doing...
16:07:39 Quit desowin ("use linux")
16:09:44Soul-SlayerHas anyone considered making an irc bot of some sort that a few people could use, where you could just do a single string, like !man or !buyguide or something, so as to get rid of all the easily solved documented questions people ask without having to spend time finding the link?
16:12:06 Quit webguest55 ("CGI:IRC")
16:13:42peturin google
16:17:47 Join miepchen^schlaf [0] (
16:18:13 Part Llorean
16:19:58 Part chrisjs169_ ("Kopete 0.12.4 :")
16:20:37 Join chrisjs169 [0] (
16:28:01amiconnre touch screens - the Cowon D2 is a touchscreen target...
16:28:29 Join saratoga [0] (i=980398fe@gateway/web/cgi-irc/
16:28:48 Join Domonoky_ [0] (
16:29:10 Nick Domonoky_ is now known as Domonoky (
16:29:14 Part Domonoky
16:29:22saratogafor GSOC2008 I'm hoping we get someone to work on mpegplayer
16:29:29 Join Domonoky [0] (
16:29:29saratogamaybe ever write an h.264 codec for it
16:29:37 Quit pepie34 (Connection reset by peer)
16:30:17low_lightinteresting...the m:robe lcd commands are similar to the archos (ssd1815)
16:33:16amiconnThe lcd command sets are rather similar in general
16:33:23amiconnIs the m:robe monochrome?
16:33:27 Join Mardoxx [0] (n=mardoxx@adsl-83-100-161-114.karoo.KCOM.COM)
16:33:56 Join TiMiD[FD] [0] (n=TiMiD[
16:34:00TiMiD[FD]hi there
16:34:44TiMiD[FD]any english native speakers here ?
16:34:46low_lightamiconn: yes. 1bit vertical
16:34:59TiMiD[FD]I've a sentence I don't understand....
16:35:28TiMiD[FD]"my friend had strictly specific that this shall me a gers only shopping trip..."
16:35:40Mardoxxthat isnt english
16:35:43Mardoxxsounds like american
16:35:46TiMiD[FD]from a singaporian guy
16:36:13linuxstbMakes very little sense to me either...
16:36:26Mardoxxmy friend has specifically told me that this will be a gers.. (germans?) only shopping trip
16:36:27TiMiD[FD]she must be on crack ...
16:36:52low_lightthis shall be a girls only...
16:37:09Mardoxxyah most likely
16:37:10Domonokyjeah, women and shopping :-)
16:37:14Mardoxxgirls += shopping ;)
16:37:19TiMiD[FD]ah that makes sense
16:37:27*linuxstb likes those kind of shopping trips - he's excluded.
16:37:41*Mardoxx likes tham and is included...
16:37:46*Mardoxx is a transvestite
16:37:48Mardoxx:O what?!
16:38:05*TiMiD[FD] hates to go shopping with girls
16:38:32*TiMiD[FD] just comes back from a 4 hours session
16:38:37Mardoxxdamn creative :'(
16:38:47TiMiD[FD]with 2 japanese freaks
16:38:54Mardoxxoh god D:
16:40:56TiMiD[FD]but I wonder why is she sending me a mail to tell me that she cannot go shopping with me ... I never asked her out in the first place ...
16:41:27amiconnlow_light: Interesting... a PP target with a monochrome screen...
16:41:51amiconnI hope that you'll adapt the grayscale lib for m:robe
16:43:25Mardoxxhow's the progress of rockbox on ZVM?
16:45:47 Quit midgey ()
16:46:55 Quit Domonoky ("Trillian (")
16:47:41 Join Domonoky [0] (
16:49:44low_lightamiconn: I will try eventually...I still need to figure out the touchpad before the thing is usable
16:50:32 Join midgey [0] (
16:51:05 Join rogelio [0] (n=rogelio@
16:51:11TiMiD[FD]what does "OEC" means ?
16:51:15 Quit rogelio (Read error: 104 (Connection reset by peer))
16:51:32 Join Lear [0] (i=chatzill@rockbox/developer/lear)
16:51:50 Join rogelio [0] (n=rogelio@
16:51:53TiMiD[FD]I've got that : "Well, maybe we shall meet again at the OEC Dorm?? ;o)"
16:52:45linuxstb"dorm" is dormitory - where students live. I'm guess OEC is the name of a college or university.
16:53:06TiMiD[FD]Osaka EC ...
16:53:32TiMiD[FD]thank you
16:54:13TiMiD[FD]but I don't want to meet her ...
16:55:56 Quit XavierGr (Nick collision from services.)
16:56:00 Join XavierGr [0] (
16:58:26 Quit midgey ()
16:58:47 Part rogelio ("Kopete 0.12.4 :")
17:00:26 Join midgey [0] (
17:01:18 Join Tha-Flash [0] (
17:02:41 Quit davina (Read error: 104 (Connection reset by peer))
17:03:17 Join davina [0] (
17:03:48***Saving seen data "./dancer.seen"
17:03:56low_lightamiconn, linuxstb: any ideas on why would the m:robe (pp5020) would get stuck in the "Initialise bss section to zero" in crt0-pp.S?
17:09:11 Quit davina (Read error: 104 (Connection reset by peer))
17:09:42 Quit midgey ()
17:10:22linuxstbSo you have a bootloader and are trying to make it load a Rockbox build?
17:10:38 Join midgey [0] (
17:11:54low_lightlinuxstb: yes. mi4 bootloader works fine. It will load OF, but gets stuck loading rockbox.
17:12:08 Quit My_Sic (Read error: 113 (No route to host))
17:12:12TiMiD[FD]low_light: just my curiosity, but on which target are you working ?
17:13:19linuxstblow_light: How do you know it's stuck in that part of crt0-pp.S?
17:13:37low_lightTiMiD[FD]: Olympus m:robe MR-100
17:14:00TiMiD[FD]I don't know this onz
17:14:09TiMiD[FD]always nice to see new ports anyway
17:14:22 Part Tha-Flash
17:15:27 Quit petur ("connection reset by beer")
17:15:36low_lightlinuxstb: I can turn the the touchpad leds on/off using the gpio ports. I narrowed it down to that loop. Leds turn on before it and are *supposed* to turn off after, but never do.
17:16:01TiMiD[FD]olympus ... I thought you were gonna make rb run on a camera ....
17:16:16Mardoxxda fuck is the point?
17:16:49 Join pepie34 [0] (
17:18:10linuxstblow_light: The only reason I can think of for the loop never ending is if the code is being overwritten somehow. i.e. that the _iram* values are incorrect. Have you looked at a disassembly of your code to see what values r2 and r3 are being initialised to?
17:20:35low_lightI will try.
17:21:51linuxstbYou could try something like "arm-elf-objdump -D -b binary -marm apps/rockbox.bin" and compare that with crt0-pp.S.
17:27:59Arathisamiconn: any progress on fixing the h10 freezes? :(
17:28:12 Quit pepie34 (Client Quit)
17:28:38 Quit printfXh4 (No route to host)
17:31:46low_lightlinuxstb: r2 = 0x5811E (_edata), r3 = 0xB1484 (_end)
17:32:07 Quit Lynx_ (Connection timed out)
17:33:29 Quit bdgraue (Remote closed the connection)
17:36:01 Join Llorean [0] (
17:37:35 Quit t0mas ("moving wifi again :)")
17:41:25 Join webguest10 [0] (i=5b7d7c27@gateway/web/cgi-irc/
17:41:36webguest10Hello, anyone there to help?
17:42:01webguest10I Have just downloaded a theme and extracted to my iPod 5th gen, how do i activate this theme?
17:42:19linuxstbMain menu -> Settings -> Browse Themes and select it.
17:42:23 Quit low_light ("CGI:IRC")
17:42:53webguest10woohoo, thats alot :D
17:43:01*bluebrother wonders if FS #5923 should get rejected
17:43:35webguest10how come it isnt like it is in the picture on Rockbox themes?
17:43:42 Join low_light [0] (i=c730180a@gateway/web/cgi-irc/
17:43:46webguest10all thats changed is the background colour =/
17:43:47linuxstbbluebrother: I thought it already was...
17:44:03bluebrotherlinuxstb: we had a patch rejected that does this for the file browser.
17:44:14bluebrotherbut I think I'll just go and reject that request.
17:44:19linuxstbwebguest10: Then your theme is broken in some way, so didn't load correctly. Most commonly you're trying a theme that requires an unofficial (modified) version of Rockbox.
17:44:43webguest10Ok, how do i delete this theme, so i can try another one?
17:45:11linuxstbYou don't have to delete it - just choose another from the Browse Themes menu.
17:45:43linuxstbDeleting it is done by deleting the files you unzipped - you need to manually do it from your computer (or from Rockbox if you're very patient)
17:45:52webguest10oh okay, thanks. il try another. On, are they not all official then?
17:47:46linuxstbRead the description underneath them - ones labelled "Any Recent Daily Build" should work.
17:48:14linuxstb(under Min Version:)
17:48:37webguest10ohhhh :) very clever, thanks alot
17:48:42 Quit midgey ()
17:48:51webguest10is it possible to do wnload those other builds easily or not worth it?
17:49:58bluebrotherthose other builds are unsupported. You can find them in the unsupported builds section of the forums
17:50:18bluebrotherbut be aware that you won't receive any support for those builds here / in the forums.
17:50:48webguest10I am extractin to the correct place yes? the /E:
17:51:08bluebrotherwell, if your player has the drive letter e: then it's e:\
17:51:31bluebrotherbtw, you can also use rbutil for installing themes.
17:51:48bluebrothercheck this:
17:52:15webguest10just 1 more, further up linuxstb said i do not have to delete the previous theme which i did not have the correct build for, if i keep it, will it harm my ipod in anyway seems as i dont have the correct build or can i just leave it?
17:52:49low_lightlinuxstb: anything odd about these values...r2 = 0x5811E (_edata), r3 = 0xB1484 (_end)
17:53:11webguest10hmm that does look good i might just download that Rockbox utility, thanks :D
17:53:57bluebrotherwell, if you have arbitrary files installed on your Ipod it might explode *g*
17:54:04bluebrotherseriously, it's just some files on the player
17:54:24bluebrotherhow should that be capable of harming the player?
17:54:33 Join midgey [0] (
17:54:52webguest10sorry i have no idea, im really dont know jack about technology/computing
17:55:17bluebrothera themes is just a bunch of text and graphics files
17:55:43bluebrothereven if you install the wrong build on your player it won't get harmed −− it just won't work properly
17:55:56webguest10ok thanks
17:56:00bluebrotherand Ipods are really hard (almost impossible) to brick at all by putting software on it.
17:56:19 Quit GodEater_ ("Leaving")
17:56:22 Join austriancoder [0] (n=austrian@
17:56:26webguest10this new theme works a treat :) thanks again
17:56:40bluebrotherif anything goes wrong you always can do a full restore using itunes.
17:56:56bluebrotherthat will everything that's on the Ipod though.
17:57:16 Quit TiMiD[FD] ("leaving")
17:57:28webguest10just while we are on the subject of itunes, is that still the way i get music on, jsut the formal way?
17:57:39webguest10normal way *
17:57:41 Join kaaloo [0] (
17:57:49 Part kaaloo
17:57:52bluebrotheryou can do that if you like
17:58:06bluebrotherthat way also the Apple firmware will recognize the files you put on it
17:58:16bluebrotherin Rockbox just enable the database.
17:58:37bluebrotheralternatively, just put the files somewhere on the player. You can just browse by filetree
17:58:49bluebrotherwhen you use itunes the files get renamed to something strange
17:59:06webguest10oh ok
17:59:21webguest10anyway i think il give this rockbox utility a go now
17:59:34bluebrotherso unless you don't need to reconizes the files you need the database in that case ;-)
17:59:59bluebrotherI can really recommend you the manual. It covers most of the initial questions
18:00:25webguest10ok thanks
18:00:49webguest10im guessing im just extacing the utility to my computer this time, rather then my ipod :)?
18:01:23bluebrotheryup, rbutil runs on your computer
18:03:09webguest10should i download the fonts package or is it not nessecesary
18:03:23linuxstblow_light: Sorry, I have to run now - I would compare them to values in the to see if they make sense.
18:03:30 Quit linuxstb ("Leaving")
18:05:21bluebrotherwebguest10: you should.
18:07:39 Quit Mardoxx ()
18:08:02 Join My_Sic [0] (
18:09:07webguest10does the rbutil tell me what builds are needed for the themes?
18:09:13webguest10i cant seem to see it anywhere
18:10:05Domonokywebguest10: the themes in rbutil come from, and this site filters them, so there should only be themes which work with the official build
18:10:18webguest10oh ok thanks
18:10:47webguest10ive noticed themes are taking longer to download/unpak with the butil, is this normal?
18:11:55Domonokycould be, but i dont know, and it doenst really matter..
18:14:59bluebrotherdepending on the graphics included a theme may be bigger than another
18:18:03 Join MonkeyTamer [0] (
18:21:22 Quit webguest10 ("CGI:IRC")
18:24:07 Quit low_light ("CGI:IRC")
18:29:52 Quit Domonoky (Read error: 104 (Connection reset by peer))
18:30:52 Join Domonoky [0] (n=Domonoky@
18:31:57 Quit Lear ("ChatZilla [Firefox]")
18:33:29 Quit obo ("bye")
18:46:10 Join _SAS_ [0] (n=sokolas@
18:46:23 Nick _SAS_ is now known as _SAS (n=sokolas@
18:47:23 Quit Tr1ckY ("reboot...")
18:48:17 Quit _SAS (Client Quit)
18:48:40 Join HellDragon [0] (i=JD@unaffiliated/helldragon)
18:49:27 Join petur [0] (n=petur@rockbox/developer/petur)
18:53:09 Join pixelma [0] (i=pixelma@rockbox/staff/pixelma)
18:54:48 Quit My_Sic (Read error: 113 (No route to host))
18:54:49 Join The-Compiler [0] (
18:58:04 Quit aliask ("ChatZilla [Firefox]")
19:03:50***Saving seen data "./dancer.seen"
19:05:21 Join gotthardt [0] (
19:06:01 Join obo [0] (
19:08:33 Join Tr1ckY [0] (
19:08:55Soul-SlayerThe wiki says that the gigabeat LCD driver is only 50% complete, is this true? It seems pretty completed to me
19:13:16 Quit Llorean ("Leaving.")
19:13:33 Join webguest92 [0] (i=42444ee1@gateway/web/cgi-irc/
19:15:18 Join Llorean [0] (
19:20:13amiconnI can now confirm that the boosting bug I observe has to do with voice and isn't codec dependent
19:20:43amiconnIf voice is enabled, the CPU stays 100% boosted regardless of how much performance the codec needs
19:20:55amiconnStays 100% boosted during playback that is
19:21:21amiconnIt only gets unboosted when stopping playback
19:21:32amiconnThis applies to all swcodec targets
19:21:52amiconnTested H180 and mini G2 now
19:22:37amiconnNo wonder I get severely reduced runtime when using the H180 in my car :(
19:23:22 Join low_light [0] (i=c730190b@gateway/web/cgi-irc/
19:23:31 Join ansivirus [0] (
19:24:15low_lightamiconn: did you see my question from earlier?
19:25:01low_lightrockbox on the m:robe gets stuck while clearing the bss section in crt0-pp.c
19:25:12 Quit webguest92 ("CGI:IRC")
19:25:16low_lightany ideas?
19:26:19amiconnWhen disabling all voice options during playback, the problem stays until stopping playback. Restarting playback is fine afterwards
19:26:27amiconnlow_light: Unfortunately not
19:26:48amiconnThis is such a basic operation...
19:27:18low_lightwould it be overwriting something important by mistake
19:29:21amiconnI'd check the .map and a disassembly whether the addresses are correct
19:29:35amiconnAlso take ram remapping into account
19:31:34low_lightThe .map and disassembly have the same values
19:31:55low_lightis that area remaped?
19:32:38amiconnWell rockbox remaps sdram to 0x0 on PP, as arm wants its exception vectors there
19:32:51 Quit ptw419 ()
19:33:00amiconnI don't know where sdram is located on m:robe after boot
19:33:36amiconnSo far, PP5020+ targets (most iPods, H10) have it at 0x10000000
19:35:31jhMikeSamiconn: voice boosts thread. playback starts. voice waits at mutex. mutex waits don't unboost cpu. :)
19:36:50amiconnBut why doesn't voice unboost when it's done?
19:37:04amiconn...before entering the mutex again
19:38:02Lloreanamiconn: But it unboosts if playback is stopped?
19:38:13 Quit Mouser_X (Read error: 104 (Connection reset by peer))
19:38:33 Join Mouser_X [0] (
19:38:37 Join toffe82 [0] (
19:39:34jhMikeSamiconn: a thread doesn't unboost until it blocks with a timeout. infinite wait don't unboost. :\
19:40:24amiconnYou're talking about auto unboost
19:40:39amiconnBut we can still use manual boost/unboost
19:41:41 Quit XavierGr (Nick collision from services.)
19:41:45 Join XavierGr [0] (
19:41:52 Join jgarvey [0] (
19:42:00amiconnAnd why does voice need to boost, btw? At least on cf, 45MHz should be more than enough for decoding 12kHz mono mp3 at < 64kbps
19:42:17 Quit Mouser_X (Read error: 104 (Connection reset by peer))
19:42:18jhMikeSwe could but if it's only only thread with a boost count it will switch a lot
19:42:35 Join Mouser_X [0] (
19:43:08jhMikeSIt always should during playback or it's really sluggish. It doesn't need it when stopped.
19:44:22 Quit Bawitdaba (Read error: 111 (Connection refused))
19:44:40 Quit petur (Read error: 110 (Connection timed out))
19:45:43amiconnVoice could keep an idle timer. If no voice for more than 1 second, unboost
19:46:29amiconnThis could still cause a quick unboost->boost sequence, but not often
19:46:56amiconnSuch a sequence could happen in other places too of course
19:47:20 Join webguest07 [0] (i=5b7d7c27@gateway/web/cgi-irc/
19:47:47 Quit spiorf_ (Remote closed the connection)
19:47:58webguest07Hello guys, i have just downloaded the theme "Theme X HighBar" and it was fine
19:48:12webguest07i then changed the font and now the display is all mucked up with different font sizes
19:48:24webguest07is their anyway to find the defauly font with this theme?
19:48:34bluebrotherwebguest07: just reload the theme
19:48:43webguest07how do i go about that?
19:48:52webguest07just choose from "brose themes"?
19:48:56webguest07browse *
19:48:58bluebrotherwell, the same way you loaded the theme the first time
19:49:38webguest07kk done, thanks
19:49:39bluebrotheralternatively you can open the theme configuration file (in /.rockbox/themes) and check for the font: line
19:49:39jhMikeSamiconn: really, I think a blocked thread need to have its boost count at least temporarily removed if blocked too long. There's other scheduler stuff in my head like a proper treatment of priority inversion so it's not an explicit thing elsewhere in the code.
19:51:05webguest07ok fantastic, thanks alot bluebrother
19:51:32amiconnDepends on how long it blocks... maybe unboost if blcoked >1 second, then reboost on unblock
19:51:49 Join My_Sic [0] (
19:51:59*amiconn still thinks all this auto-unboost and priority stuff made the scheduler more complex than it should be
19:52:28*amiconn also wonders what causes the hard freezes when browsing a lot with voice enabled
19:52:28 Quit webguest07 ("CGI:IRC (EOF)")
19:52:51amiconnHad to use paperclip equivalents 3 times this week...
19:54:12jhMikeSpriority inheritance should be simple to implement. I just need a few details worked out. Keeping interrupts out of the scheduler would simplify things a whole lot. That's part of the core patch but perhaps should be done separately.
19:54:51amiconnKeeping interrupts out of the scheduler?
19:55:16jhMikeSwhen thet post messages, they wake the waiting threads
19:56:06amiconnYes, as they should afaiu
19:56:24 Join desowin [0] (
19:56:37jhMikeSthere's all kinds of elaborate stuff in there to keep it atomic which can go if interrupts post to a system queue instead.
19:57:33*amiconn can't follow it seems :(
19:58:54amiconnIsn't it just a quick "disable irq - post message - enable irq" sequence?
20:00:02 Quit low_light ("CGI:IRC (EOF)")
20:02:29jhMikeSno, there's a lot to keep the scheduler locked well inside the code and also to wake from interrupt posts on the next tasks switch or a wake from a mutex can corrupt the lists
20:03:48jhMikeSso is the origin of funky functions like set_irq_level_and_block_thread_w_tmo and such
20:04:09pixelmaa hard power-off on sansa is "holding the power button for at least 15 seconds" and should always work... right?
20:04:33*amiconn really can't follow now :(
20:05:41*jhMikeS supposes a close following of the scheduler and kernel object code would do the trick and maybe bring other ideas?
20:06:10amiconnI have given up following it closely when those dynamic lists were introduced
20:06:24amiconnThe single thread array was nice and simple, and it worked
20:06:39amiconnThe priority stuff makes up for further headaches
20:07:21amiconnThis is way more complicated than RE-ing some arm code imo
20:07:22Lloreanpixelma: Yes
20:07:34jhMikeSthe priority stuff wasn't given a treatment for inversion which will bite back without it
20:07:54pixelmaLlorean: thanks, just wanted to give the correct info
20:08:38*amiconn admits that he doesn't have experience with thread programming other than rockbox
20:08:53 Quit My_Sic (Read error: 104 (Connection reset by peer))
20:08:55jhMikeSI'll have to have a look at the old threading and see where things weren't entirely IRQ safe. I suspect the queue race condition existed then.
20:09:31amiconnWhat queue race condition?
20:09:52amiconnInterrupts are disabled during queue posts
20:09:54jhMikeSWhere a thread could get abandoned.
20:10:07jhMikeSthey had to be disabled during queue_waits too
20:10:10amiconnThe old threading was simple round robin
20:10:24amiconnNo thread ever got abandoned
20:11:07amiconnCertainly not during wait
20:11:22amiconnOnly during the actual queue read when a message arrived
20:11:43jhMikeSan interrupt could post a message but the thread hasn't gotten to waiting. the thread about to wait on the empty queue (as it saw it at the time) misses the wakeup.
20:12:10amiconnThe old threading *always* woke up *all* threads
20:12:13 Join My_Sic [0] (
20:12:36amiconnThe thread was responsible for checking and going back to sleep when there was nothing to do
20:14:01 Join moos [0] (
20:14:02amiconnSo the maximum possible message delay was once tick
20:14:35jhMikeSSo a task switch over the entire list occurred every time. Now they just stay blocked an take no CPU time.
20:15:12 Quit Nico_P (Remote closed the connection)
20:15:21Slasheriyes, the old threading was not that performance efficient
20:16:21amiconnI think the difference is negletible
20:16:43amiconnSure, it went through the whole list instead of just the running threads
20:17:12amiconnBut how many threads are there? About a dozen. ANd how often does a thread switch occur?
20:17:19jhMikeSI think as rockbox grows more threads it will matter more.
20:17:54amiconnAnd the sleep check was just moved into the scheduler instead of having it in the threads. The difference being being one function call
20:17:54jhMikeSpretty often actually
20:18:47amiconnThe scheduler now needs to check whether to wake up a thread. More complex and error prone than just waking all and letting them do the check
20:19:28amiconnThe priority stuff might be necessary on swcodec though to avoid too complex checks in the threads themselves
20:19:34jhMikeSyou mean for the timeout list?
20:20:18Slasheriamiconn: for blocked threads there are no checks at all, those are removed from scheduling lists
20:20:29jhMikeSat least these check require no switch to do now.
20:21:30amiconnSlasheri: Some code certainly needs to unblock these when necessary?
20:21:53 Quit maffe (Remote closed the connection)
20:21:57Slasheriamiconn: indeed, but that kind of threads can remaing blocked for quite a long time
20:22:21amiconnsleep(1) ??
20:22:38Slasherino, that's timeout state
20:22:48Slasheriblocked is mutex or queue_wait
20:23:20amiconnThe main thread processing repeat button events...
20:24:28 Join webguest44 [0] (i=5b7d7c27@gateway/web/cgi-irc/
20:24:38webguest44Hi, i was wondering if anyone could help me
20:24:58webguest44I have instealled rockox on my 5th gen Ipod video and was wondering how to access my video
20:25:07webguest44all i can find is my music but not my movies
20:26:17Soul-SlayerYou can only play .mpg files from within Rockbox, which have been dragged and dropped onto the drive in Disk Mode
20:26:31Soul-SlayerTo play videos you put on in itunes you need to use the appleOS
20:26:55webguest44how do i change back to it?
20:27:06Soul-SlayerWhen you turn your ipod on, flick hold on
20:27:20Soul-SlayerWhen it starts booting into the appleOS, take hold off again
20:27:26Soul-SlayerReset the player to return to rockbox
20:28:04webguest44Ingenious :) thanks
20:28:32 Quit webguest44 (Client Quit)
20:28:35jhMikeSamiconn: button evens are a very long block
20:30:02amiconnUp to 20 events/sec
20:30:26 Join lee-qid [0] (
20:31:02 Join maffe [0] (
20:31:33jhMikeSthreads can run again much sooner than that
20:32:24 Quit iamben (Read error: 104 (Connection reset by peer))
20:35:29 Join spiorf [0] (
20:38:36 Quit ansivirus ("Leaving")
20:47:07 Join low_light [0] (i=c730190b@gateway/web/cgi-irc/
20:50:03 Quit My_Sic ("Miranda IM! Smaller, Faster, Easier.")
20:53:04 Quit Mouser_X (Read error: 110 (Connection timed out))
20:53:40 Quit Arathis ("Bye, bye")
20:54:15*Domonoky reads about a official Theme repository in the forums.. any changes to rbutil needed for this ?
20:55:20saratogaBagder: can you submit my GSOC evalution?
20:55:31saratogai think they're supposed to be in soon
20:55:38 Join PaulJam [0] (
20:55:49saratogapreglow: unless you're actually here?
20:57:03 Join BK [0] (n=Non@
20:57:31moossaratoga: don't worries you made a good job ;)
20:58:31 Quit Rick (Read error: 104 (Connection reset by peer))
20:58:42Domonokywhoa gsoc midterm evaluation.. are all students on track ? :-)
20:58:43moosNico_P is in vacations, no?
20:59:23 Join Rick [0] (
20:59:44MonkeyTamerWell, there's still the final
20:59:58scorcheDomonoky: not sure yet...should just be a few minor changes, but as i said when i first started working on it...i will let you know when it is ready to go live =)
21:00:09 Join Arathis [0] (
21:00:14scorchesaratoga: he is on vacation and isnt really around
21:00:30saratogawho are the GSOC rockbox representives then?
21:00:41mooswho is your mentor?
21:00:43saratogaI think anyone who is registered with google can submit it
21:00:51Domonokyscorche: good, tell me if any changes are needed, now at the weekend i have a bit time :-)
21:00:52saratogapreglow, however i believe he is on vaction
21:00:53pixelmascorche: didn't you post a link here regarding the midterm evaluation?
21:01:06pixelmamaybe a week ago
21:01:07scorcheas i said before, i *think* Llorean was boosted to admin, but i am not sure what that entials
21:01:22scorchepixelma: perhaps...i dont remember me doing that though
21:01:36moossaratoga: since linuxstb help you on your project maybe he is the best one to make your "evaluation" ?
21:01:50Lloreanscorche: Yes, I can fill them out
21:02:09scorcheLlorean: then you might want to talk with saratoga =)
21:02:10pixelmacould be I'm confusing you with someone else... trying to find it in the logs
21:02:43 Join lostnihilist [0] (
21:03:07moosis someone know what the "Clear recording directoryé
21:03:17moosdirectory" is supposed to do?
21:03:32saratogalooking more closely, it seems we have until Sunday to hand them in
21:03:36saratogaso i'll try preglow again
21:03:46Lloreansaratoga: If you don't get in contact with him by the due date, I can fill it in.
21:03:53***Saving seen data "./dancer.seen"
21:03:55saratogaLlorean: thanks
21:03:55moosjdGordon seems to doesn't like the "desc:" entry in the .lang file
21:05:01MonkeyTamerhow does the grading or evaluating work anyways? Is there some sort of incentive for people keeping on track other than intrinsic motivation?
21:05:28DomonokyMonkeyTamer: money :-)
21:05:38MonkeyTameroh, that explains it then
21:05:42pixelmamoos: you can set any directory as recording directory now (it's in the context menu on a directory). Later you can clear this setting to default to the root again (where your recorded file will end up then)
21:06:15scorcheDomonoky: dont forget the shirt!
21:06:25moospixelma: ah ok I got it now, merci
21:06:38Domonokyscorche: yeah, shirts for those where mony doesnt count :-)
21:07:19pixelmamoos: similar to "set as backdrop"/"clear backdrop"... your welcome :)
21:07:51moosyeah that was the "clear" that confused me :)
21:08:33pixelmadelete the entire recording directory! ;)
21:10:48jhMikeSdoes the samsung tuner really need the wait before being unpaused/started or just the philips? sanyo doesn't.
21:12:45bluebrotherhmm. Why doesn't the recording dir default to /recordings as it did before?
21:13:58bluebrotherand maybe "reset" instead of "clear" would be clearer?
21:16:40pixelmanot sure but I believe it has to do with the fact that you wouldn't be able to set the root as recording directory...
21:16:49moosbluebrother: I used reset for french .lang
21:17:04moosthat what I thought when pixelma explained me
21:17:23pixelmabecause you can't call the context menu on your root directory
21:26:02bluebrotherbut "clear" sounds like it will delete everything in the recording folder.
21:26:18moosthta what I thought first too :)
21:26:56moosthe term isn't clear, and even no description on the /lang file for translators...
21:27:28bluebrotherI also somewhat dislike that you can't query the current set directory.
21:28:33moosbut I guess IdGordon spoke about those changes before to commit? :)
21:29:32pixelmascorche: it was markun... sorry :) But this is what I remembered reading
21:30:09bluebrotherthat "write .cfg file" could also better named as "write recording preset" IMO
21:30:47bluebrotherwould make it not only clearer but also easier to explain in the manual.
21:31:35pixelmabluebrother: maybe it's only a reused string
21:32:17bluebrotherpixelma: well, that's possible −− but nothing that can't be changed ;-)
21:32:26*moos try to always be the more teaching as possible for new french users
21:32:26pixelma(makes it "cheaper" if you look at binary size for now)
21:32:51bluebrotherdoesn't lang files have their own buffer anyway?
21:33:00 Quit jgarvey ("Leaving")
21:33:39moosone that capable to fit the bigest .lang IIRC
21:33:47moosgreek or something
21:33:53 Quit MonkeyTamer ("ChatZilla [Firefox]")
21:34:27 Quit donutman25 ("ChatZilla [Firefox]")
21:34:32*moos go away to enjoy the sun a bit more ;)
21:34:56pixelmaI was talking about binary size and every added string to english.lang increases it - that's why the langV2 cleanup..
21:35:26bluebrotheradded strings add to the binary size? You sure?
21:36:05bluebrotheras far as I understand langv2 is to have target dependent lang files, so the lang buffer can get decreased depending on the number of strings for a target.
21:37:21 Join stripwax [0] (
21:38:19pixelmabinary size... for example langV2 cleanup helps dropping that by 2500 bytes for my OndioFM - and that's the reason why the recording dir patch increased binary size a bit for all targets also the ones that can't record the way it is atm
21:40:52pixelmaonly changes to english.lang though as it's build-in so you at least have the english string even if your language file isn't up to date
21:41:07pixelma(as I understand it)
21:43:09Lloreanbluebrother: English is compiled into Rockbox.
21:44:37 Join petur [0] (n=petur@rockbox/developer/petur)
21:45:39 Quit XavierGr (Nick collision from services.)
21:45:43 Join XavierGr [0] (
21:46:59pixelmaLlorean: thanks for putting my chaotic explanations in a simple form :)
21:47:37Lloreanpixelma: Honestly I just missed that bit in your explanation. A wee bit scatterbrained again
21:48:01pixelmame, you, both?
21:49:01pixelmame too = both
21:49:59 Quit midgey ()
21:50:02bluebrotherLlorean: ah, haven't thought of that.
21:52:23 Quit stripwax ("Miranda IM! Smaller, Faster, Easier.")
21:54:57Soul-SlayerHas anyone started working on a seek feature for mpegplayer yet, or is that on the back burner for a while?
21:55:07Soul-SlayerPresuming it's planning on being implemented
21:57:44 Join Domonoky_ [0] (
21:57:44 Quit Domonoky (Read error: 104 (Connection reset by peer))
21:58:07peturjhMikeS: around?
22:00:16 Join bdgraue [0] (
22:01:00 Join My_Sic [0] (
22:02:25 Quit SirFunk (Connection timed out)
22:03:42 Quit z35 (Read error: 113 (No route to host))
22:05:35jhMikeSpetur: sitting here wasting time on BOS
22:06:18 Join davina [0] (
22:06:23peturjhMikeS: I'm looking at the report about a full disk during recording
22:06:50jhMikeSthe corruption one?
22:06:56peturI don't quite understand enc_events_callback()
22:07:09peturyes, file corruption
22:07:34*jhMikeS puts on recording cap
22:08:01peturthe last line in that function is never reached, correct?
22:08:02jhMikeSthat's the file split event callback
22:08:30peturwriting is done in on_write_chunk()
22:08:36jhMikeSyes, it is if the called functions fail
22:09:20jhMikeSif (function()) then return; <= else it drops through to the end
22:09:51*jhMikeS is looking in wav_enc.c
22:09:53peturthat last else catches everything, no?
22:10:09peturhmmm right, wrong file
22:10:19 Join midgey [0] (
22:10:28*petur was looking at mp3_enc.c and doesn't know why
22:10:56jhMikeSthat one looks ok too
22:11:18peturstill, that last else catches everything, no?
22:11:39jhMikeSeverything it doesn't care to process
22:12:04jhMikeSif it takes no action it has no reason to fail it
22:12:34 Join z35 [0] (
22:12:52peturwell if writing fails, on_write_chunk returns false, and eventually the function returns in the last else clause
22:13:48peturso CHUNKF_ERROR is never set
22:14:37peturdoesn't explain file corruption though
22:14:39jhMikeSit jumps to the end, sets the flag and returns. it's been tested and failing a write in there will set it.
22:15:00 Join ompaul [0] (n=ompaul@freenode/staff/gnewsense.ompaul)
22:15:35peturhow does it evade that last else { return; } ?
22:17:04peturthat if() else if()... else construction handles every possible code path, that last line can never be reached
22:18:46 Part BK
22:20:09jhMikeSif (event == ENC_WRITE_CHUNK) the return gets executed if the function succeeds. the else clause isn't reached. Perhaps a redo using a switch statement equivalent to illustrate?
22:20:49peturthe else clause is reached if it fails....
22:21:36jhMikeShow? I don't understand how you see that.
22:21:38peturmy problem is: after that function, any info on the fact that the operation failed is lost (imho)
22:23:55peturyes, that code would be correct, it is _not_ equivalent to what is in there now, agree?
22:24:22 Nick chrisjs169 is now known as chrisjs169http (n=ubuntu@unaffiliated/chrisjs169)
22:24:50 Nick chrisjs169http is now known as chrisjs169 (
22:25:33peturshall I change that and test?
22:25:52jhMikeSno, I don't but feel free to compare
22:27:03jhMikeSonce it returns, the error flag is checked in pcm_record.c which aborts it
22:27:44jhMikeSis_file_data_ok in the encoder checks the flag before doing anything as well
22:27:52peturbut the flag is _never_ set
22:28:22peturthat last else { return;} catches everything....
22:28:27jhMikeSI've returned false deliberately from those to test and recording stops as it should
22:28:44jhMikeSnot if one of the other clauses is entered
22:29:11peturow sh*t
22:29:14 Quit The-Compiler (Read error: 104 (Connection reset by peer))
22:29:24peturI see now... sorry
22:29:36*petur runs after forehead and slaps hard
22:29:51 Quit moos (Read error: 110 (Connection timed out))
22:30:02jhMikeSactually I'm wondering what happens if ATA is asked to write four sectors and only say, 2 are left free on the drive
22:30:10 Quit z35 (Read error: 113 (No route to host))
22:30:14*scorche thinks forehead should file for separation with all the abuse it gets
22:30:50peturit will return the number of bytes written, and enc_write_chunk returns false because the amount written is incorrect
22:34:16jhMikeSwill it catch the overflow on the volume?
22:34:37jhMikeSthe ata driver I'm saying
22:35:05jhMikeSif it's _told_ the incorrect amount was written, yes it returns false
22:36:33*jhMikeS notices a funny typo: "called version often - inline" :)
22:38:13pixelmais that an error that occurs always with swcodec recording and a full disk or only sometimes? Because I've never had problems with it on my Ondio (just got the "disk full" splash)
22:38:41peturI have never managed to fill my disk :/
22:39:20pixelmawell that's easier with 128MB of storage internally (though recording as mp3 only) ;)
22:39:35*petur has 30GB free
22:39:49 Join amiconn_ [0] (n=jens@rockbox/developer/amiconn)
22:40:16pixelmawith a 80GB disk...
22:40:54 Quit otih_ (Remote closed the connection)
22:42:29*pixelma could quite easily try that out on her M5 now (66,4 MB free)
22:43:18pixelmathough I can only record from the internal memory atm (no subpack)...
22:45:07jhMikeSMost of the time the problem is clearly caught (see forums) but something in the file apis fails to tell recording the write (should have) failed on a rare occasion :\
22:45:13 Part Llorean
22:46:42jhMikeSWith recording being the biggest user of writing to disk I'd expect any problems with writes to show most often when using it.
22:51:59pixelmaso maybe hwcodec recording could have the same problem, just that I haven't encountered it myself
22:56:35 Quit amiconn (Read error: 110 (Connection timed out))
22:56:36 Nick amiconn_ is now known as amiconn (n=jens@rockbox/developer/amiconn)
22:57:10*jhMikeS wonders about mostly filling a disk in advance and running recording over and over until the problem hits
22:58:05Soul-SlayerWhy not just disallow recording if there isn't atleast 1mb of space remaining or similar?
22:58:28Soul-SlayerToo restrictive?
22:58:33 Quit Hadaka (
22:58:33 Quit krp (
22:58:50NJoinHadaka [0] (
22:58:50NJoinkrp [0] (
22:59:47peturwhat if you're douing a 3 hour recording to WAV?
23:00:06saratogawould someone mine taking a quick look at my ARM assembly fixed point Multiply Accumulate function?
23:00:19saratogaand tell me if i got the gcc syntax right and all that?
23:00:27saratogaits just 3 lines of ASM
23:00:32 Join stripwax [0] (
23:00:47Soul-SlayerWell, have it cut off once it hits 1mb or so... Surely using up every last byte on the drive is going to cause problems anyway, saving config files and similar
23:01:55 Quit freqmod (Remote closed the connection)
23:01:59 Quit krp (
23:01:59 Quit Hadaka (
23:02:01 Join jgarvey [0] (
23:02:24NJoinHadaka [0] (
23:02:24NJoinkrp [0] (
23:03:18 Part low_light
23:03:21peturI really suspect that filling those last clusters will not work out well as they will be scattered around the drive so writing speed will be very slow
23:03:23jhMikeSI need a utility to logically fill the disk to a certain point without actually writing all that data :\
23:03:48peturor stuff it before testing
23:03:54***Saving seen data "./dancer.seen"
23:04:42 Quit midgey (Read error: 104 (Connection reset by peer))
23:05:32jhMikeSstuffing that much takes a long time too...of coure I wouldn't actually record so much. it just needs to hit the wall. using recording probably isn't needed since disk full can be faked to tests its response to an error return.
23:06:00Bagderyou can use dd to create huge files
23:06:37jhMikeSBagder: without actually writing all those bytes?
23:07:07BagderI guess you could make a smaller partition
23:12:16 Join bospaadje [0] ( long as everything behaves the same as for one big one
23:13:43*petur needs another unit to mess with...
23:13:46 Quit bdgraue (Remote closed the connection)
23:14:41 Join bdgraue [0] (
23:15:50jhMikeSpetur: noone with the problem ever said they had multiple partitions or anything? or all?
23:16:08 Quit davina ("xchat on Ubuntu 7.04")
23:16:09peturnot that I remember
23:17:35 Quit bdgraue (Remote closed the connection)
23:19:04jhMikeSthis sort of problem generates loads of speculation...most of it wrong :\ just like that other one. rrr.
23:19:31peturjhMikeS: upon reading the report (5852) again, it could just be that the header wasn't written. Thus creating a wav that looks like it is bad
23:21:54jhMikeSa disk truely full I suppose could do that. just closing a file can require a sector free to fsync it.
23:22:32 Join inversions [0] (
23:22:42 Join bdgraue [0] (
23:22:56peturyou mean writing a cached cluster?
23:23:00jhMikeStrying to write the header might try to finish off something at the end.
23:23:08peturI see
23:23:37jhMikeSyes. it has to wait for a full one in order to write, correct? recorded data isn't multiples of 512.
23:23:57jhMikeSnot all of it anyway
23:26:26 Join z35 [0] (
23:27:42jhMikeSpetur: one person iirc said recording just kept going and never reported an error. didn't you link me to that?
23:28:29peturI can remember something like that but no details... can't remember it was me who told you
23:29:15jhMikeSI think you posted them in the forum
23:29:16peturhmmm when writing the tail bytes, it does a read of the next sector so the close should work as we don't cache more than a sector
23:29:54 Quit saratoga ("CGI:IRC")
23:30:34 Join lymeca_ [0] (n=lymeca@unaffiliated/lymeca)
23:30:42 Nick lymeca_ is now known as lymeca (n=lymeca@unaffiliated/lymeca)
23:31:04lymecaI have a 30GB 4th Gen iPod that is HFS+ formatted that I need to switch to FAT32 in order to put ROckbox on it.
23:31:13jhMikeSpetur: and if there's no next sector?
23:31:24lymecaHowever on the wiki page for converting to FAT32 there is no partition table download for 30GB 4th Gen
23:32:18peturthen it returns negative or the number of bytes it did write
23:32:33*jhMikeS supposes recording could try to write the header for everything it did write so far even if an error happens
23:32:56peturthat would be nice, yes
23:33:36jhMikeSmight save a few more files from needing a repair job
23:35:38Soul-Slayerlymeca, read the note above it
23:35:43Soul-SlayerAbove the table
23:36:03Soul-SlayerOh wait, you have a 4g, sorry, only looked at the '30gb' :$
23:37:37Soul-SlayerI can't help there, sorry.
23:41:53*petur selects the option 'early night'
23:41:57 Quit petur ("zzzz")
23:43:28 Quit Arathis ("Bye, bye")
23:44:43 Join midgey [0] (
23:47:37 Join saratoga [0] (i=9803c6dd@gateway/web/cgi-irc/
23:52:10 Quit midgey ()
23:52:51saratogain arm assembly, the shift field always refers to the second operand right?
23:53:20 Quit jgarvey ("Leaving")

Previous day | Next day